Respiration Physiology. 82 (1990) 295-306 Elsevier
295
RESP 01719
Action of the intercostal muscles on the rib cage A.F. DiMarco, J.R. Romaniuk* and G.S. Supinski Department Of Medicine. MetroHealth Medical Center and Case Western Reserve University, Cleveland, Ohio 44109, U.S.A. (Accepted 24 July 1990) Abstract. Recent studies suggest that the parasternal muscles (PA) are primarily responsible for rib cage expansion during eupneic breathing with a much lesser role played by the interosseous external intercostals (EI). The purpose of the present investigation was to assess the capacity of the El to expand the rib cage during spontaneous breathing in the absence of coincident ipsilateral PA activation. In 9 anesthetized dogs, we measured PA EMG and length in the 3rd interspace and El EMG and length in the 3rd and 4th interspaces. During resting breathing, each muscle was electrically active and shortened to a similar degree, approximately 3 ~o of resting length. Following ipsilateral PA denervation ( I st through 6th interspaces), the level of El shortening in the 3rd and 4th interspaces was maintained, but with an increase in neural drive to these muscles. The parasternal muscle in the 3rd interspace lengthened during inspiration. Subsequent sequential dencrvation of El in the 3rd and 4th interspaces resulted in their lengthening. In 4 additional animals, axial motion of the 4th rib was measured in the mid axillary line. Ipsilater',d PA denervation had no significant effect on rib motion. External intercostal denervation (3rd interspace), on the other hand, had a substantial impact on rib motion, causing the 4th rib to move in the caudal direction during inspiration. Our results indicate that: (a) the EI of the lateral rib cage are capable of elevating the ribs during inspiration independent of PA contraction; (b) PA contraction contributes to El shortening during eupneic breathing and (c) regional loss of muscle activation results in local rib cage distortion, suggesting that the upper rib cagc has multiple degrees of freedom.
Animal, dog; EMG, of respiratory muscles; Respiratory muscles, parasternal intercostals, interosseous external intercostals; Rib cage, movement and intercostal muscles
Both the parasternal (PA) and external interosseous intercostal (El) muscles of the upper rib cage are electrically active and shorten during inspiration (D'Angelo, 1982; Decramer and DeTroyer, 1984; DeTroyer and Ninane, 1986; Decramer et al., 1986). There is some controversy, however, concerning the relative importance of these two muscle groups in producing rib cage expansion. Some studies have suggested that the PA muscles are the primary agonists producing rib cage expansion with a much lesser Correspondence to: A.F. DiMarco, Metrollealth Medical Center, Pulmonary Division, 3395 Scranton Road, Cleveland, OH 44109, U.S.A. * Present address: Dept. of Ncurophysiology, Polish Academy of Sciences, Medical Research Center. Warsaw, Poland. 0034-5687/90:'S03.50
(~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
296
A.F. DIMARCO et al.
role played by the EI muscles. In fact, recent studies suggest that the interosseous intercostal muscles may subserve more important postural, as compared to ventilatory, functions (Decramer et al., 1986). This contention is based, in part, upon the findings that there are much greater changes in interosseous intercostal muscle length associated with rotation of the trunk compared to that associated with respiration. Other studies, however, have demonstrated that the El, like the PA muscles, are capable of producing large inspired volumes and changes in airway pressure (Budzinska et al., 1989), suggesting an important ventilatory role. These latter studies, however, were somewhat artificial in that they were performed utilizing electrical stimulation techniques and under conditions of near maximal intercostal muscle activation. The purpose of the present study, therefore, was to assess the capability of the external intercostal muscles to elevate the ribs during their spontaneous activation. It the external intercostals are, in fact, capable of elevating the ribs while contracting alone, their level of shortening, an index of rib elevation as well as axial motion of the ribs, should be maintained in the absence of parasternal contraction. External intercostal EMG and length and rib motion was assessed, therefore, both before and 'after denervating the ipsilateral PA musclcs in anesthetized dogs during resting breathing. To separate that portion of the El length changes and rib motion due to their own motor activation from that caused by other contracting inspiratory muscles, measurements were also made following subsequent EI denervation.
Methods
Experiments were performed in 13 supine dogs whose weight varied between 16 and 20 kg. Animals were anesthetized with pentobarbital sodium (Nembutal) at an initial dose of 25 mg/kg given intravenously. Supplemental doses (2-4 mg/kg) were given as needed to maintain a stable anesthetic level, i.e. absent response to noxious stimuli, but titrated to maintain an intact corneal reflex and normal end-tidal Pco2 throughout the experiment. A large bore endotracheal tube (ID = 9 mm) was sewn into the trachea in the mid-cervical region. An arterial catheter was placed in the femoral artery for blood pressure monitoring (Statham P23) and intermittent blood gas analysis. Another catheter was placed in the femoral vein to administer IV fluids and additional doses of the anesthetic agent. Changes in end-tidal Pco, were monitored at the tracheal opening via a rapid responding CO2 analyzer (Beckman LB-2). Body temperature was maintained at 38 + 0.5 °C by means of a homeothermic blanket (Harvard Apparatus, Cambridge, MA). Tidal volume was obtained by electrical integration of the flow signal from a pneumotachograph (Fleisch 1), connected to a pressure transducer (Validyne MP 45). Tracheal pressure was measured with a separate differential pressure transducer (Validyne MP 45). All signals were recorded on a 12-channel Electronics for Medicine oscilloscopic recorder. The superficial muscles of the rib cage including the pectorals, transvcrsus costarum and portions of the serratus anterior and scalenous muscles were sectioned unilaterally
INTERCOSTAL MUSCI.E ACTION
297
to facilitate access to the intercostal musculature. Bipolar stainless steel electrodes were then implanted under direct vision into the parasternal muscle (PA) of the 3rd intercostal space (approximately 2 cm lateral to the sternum) and into the external intercostal (El) muscles of the 3rd and 4th intercostal spaces (approximately 1 cm dorsal to the midaxillary line). Mass efferent activities of each muscle were amplified, rectified and processed by RC circuits containing third order low pass filters with a time constant of 100 msec (Charles Ward Enterprises, Ardmore, PA) to provide signals approximating moving averages of the recorded activity. Parasternal and El muscle length was measured in the same interspaces by sonomicrometry (Newman et al., 1984). Pairs of piezoelectric crystals (Model 120, Triton technology, San Diego, CA) were implanted into the PA muscle just lateral to the sternum and into the El near the mid-axillary line, just medial to the recording EMG electrodes. To facilitate placement of the piezoelectric crystals in the EI muscles, the animal was rotated to the opposite side and subsequently returned to the supine posture. Changes in intercostal muscle length were expressed as a percentage of resting length (~o LR). Similar to previously described methods (DeTroyer et al., 1985), a hook was screwed into the 4th rib in the mid-axillary line. An inelastic thread attached to the hook was passed over a pully connected to an appropriately oriented linear displacement transducer (Model 33-03-981, Brush Instruments, Cleveland, Ohio) to measure axial rib motion. Resting length of the intercostal muscles and rib position were determined by hyperventilation-induced apnea. In some animals, it was also necessary to administer small doses of succinylcholine to achieve complete apnea. Protocol. In 9 animals, baseline measurements of muscle shortening and peak EMG were made for each intercostal muscle and repeated after each of the following maneuvers. First, the internal intercostal nerve in the 3rd interspace was sectioned to denervate the parasternal muscle in this interspace. Subsequently, the internal intercostal nerves of the ipsilateral 1st, 2nd, 4th, 5th and 6th interspaces were sectioned just lateral to the costochondraljunctions to eliminate PA action over the ipsilateral rib cage. Denervation of the PA was limited to the ipsilateral rib cage since it was felt that these muscles have the greatest influence on the ipsilateral rib cage motion during resting breathing. Finally, the El nerves of the 3rd and 4th interspaces were sequentially sectioned. Following each muscle denervation, a 10-15 rain recovery period was allowed to elapse before measurements were repeated. Complete denervation of the PA (3rd interspace) and El muscles (3rd and 4th interspaces) was confirmed by the absence of inspiratory EMG activity in each interspace. An average of 3 breaths were taken under each condition to determine the level of E M G and muscle shortening. In 4 separate animals, axial motion of the 4th rib and El EMG were monitored under control conditions, following ipsilateral PA dcnervations (lst through 6th spaces) and subsequent El (3rd space) denervation. An average of 3 breaths was taken to determine the level of EMG and rib motion. The effect of denervation of each muscle was asscssed statistically by paired t-tests.
298
A.F. DIMARCO et al.
Statistical comparison between the 3 muscles were made by employing a repeated measures analysis of variance and p o s t hoc t-test.
Results
Mean inspired volumc and end-tidal Pco~ were 350 ml _+ 23 SE and 40 mmHg + 1 SE, respectively, under control conditions and were not significantly affected by ipsilateral PA and EI denervation, lnspiratory electrical activity and active inspiratory shortening was present in each of the 27 intercostal muscles studied. The effects of initial denervation of the parasternal muscles followed by El muscle denervation is shown for a representative animal in fig. 1. The PA of the 3rd interspace and EI of the 3rd and 4th interspaces were electrically active and shortened during inspiration. An initial rapid shortening to resting length is followed by more gradual shortening below resting length. The initial rapid phase of shortening has been described previously for the PA and is attributable to expiratory muscle relaxation (Ninane et al., 1986). The subsequent shortening below resting length during inspiration is the subject of this study and is referred to as active shortening. This phase of shortening occurs in synchrony with the onset of intercost',d EMG activity. Denervation of PA of the 3rd space resulted in the elimination of active PA shortening during inspiration. As shown in this example, PA denervation had little effect on the level of EI activation or degree of inspiratory shortening of the EI in the 3rd interspace. There was a small increase in EI shortening in the 4th interspace with essentially no change in its EMG. Following subsequent denervation of the other ipsilateral PA muscles (lst, 2nd, 4th, 5th and 6th interspaces), EI EMG of both the 3rd and 4th interspaces increased. This increased EMG activity, however, was not associated with an appreciable change in peak inspiratory EI shortening. Subsequent denervation of EI muscle of the 3rd interspace resulted in inspiratory lengthening of this muscle and a small increase in inspiratory shortening of El in the 4th interspace without change in the level of its motor activation. Subsequent denervation of EI in the 4th space resulted in inspiratory lengthening of this muscle as well. The level of active inspiratory shortening under control conditions for the PA (3rd interspace) and El (3rd and 4th interspaces)was 3.3 ~o LR +_ 0.6 S E, 3.4~o LR +_ 0.5 S E and 2.6~, LR + 0.3 SE, respectively, and not significantly different from each other. The effects of initial PA denervation (3rd interspace) on PA shortening is shown for each animal in fig. 2. In each animal, there was a marked reduction in active PA shortening. Mean PA length (3rd interspace) changed from an inspiratory shortening under control conditions to + 0.7% LR +_ 0.2 SE following PA denervation. PA denervation resulted in a small increase in EMG to the El in the 3rd and 4th interspaces (NS), as shown in fig. 3 and no significant change in El length. Denervation of the PA (lst, 2nd and 4-6th interspaces) resulted in further lengthening of the PA (3rd interspace) to + 1.1 ~o LR + 1.2 SE (NS). Mean El EMG of the 3rd and 4th interspaces increased further following ipsilateral PA denervation (lst through 6th interspaceXP < 0.01 compared to
299
INTERCOSTAL MUSCLE ACTION PoM Palaste~r~| IC D e r ~ f v a l i o n (3rd space)
Control
P,last¢Inal
Po~t P a l a s l e r ~ l
Posl Exter~,al
~ n , ervalion
IC [~,,efvatioqrl (3rd ~ ( . e )
(1-6 spaces)
Pos! I xternal IC [ ~ e r v a t l o n (41h space)
0
(a,uP
Ixt~nal K:
lao)
lo E
o
~ - r ~ - ~ , - ~
length I3ld sp~e) (%tR)
-s
I%I~)
.(,
~
--"--"'--~-- ~
-
- ~=
L
\
[ ,le,.JI IC Ir:,glh t41t, ~l,~e p I%IR )
4
' ,
.-
/
~
~\ '
:\/
'k./
.S'
Fig. 1. Effect of sequential denervation of the parasternal (PA) and external intercostal (El) muscles in a single animal. Initial PA denervation in the 3rd interspace eliminated active shortening of that muscle. Subsequent denervation of the other ipsilatcral PA (lst through 6th interspaces) resulted in an increase in El muscle activation but no significant changes in active El shortening. Subsequent El denervation in the 3rd and 4th intcrspaces resulted in El lengthening during inspiration. See text for further description.
-9
-6 Parasternal IC
length (3rd space)
-3
(%LR)
Post Denervation 0
0
0 00
O0 0 0
+3 0
I
1
I
-3
-6
-9
Parasternal IC Length Control (%LR) Fig. 2. Identity' plot relating active parasternal intercostal shortening in the 3rd interspace before (X-axis) and after (Y-axis) its dencrvation. Dencrvation of this muscle virtually eliminated its inspiratory shortening.
300
A.I.. DIMARCO 200
3rd space
et al.
4th space
100 x
Control F77";I Post Parasterna: Denervation (3rd space) Post Paras:err'al Denervation (1-6 space) t"ig. 3. Effect of parasternal (PA) denervation on peak external intercostal (El) E M G in the 3rd and 4th
interspaces. Denervation of the PA muscle in the 3rd interspace resulted in a small increase in El E M G (NS). Subsequcnt dcnervation of thc ipsilateral PA intercostals in the 1st through 6th spaces resulted in significant incrcases in pcak EI EMG in both the 3rd and 4th interspaces.
control values). The mean level of active inspiratory shortening o f the EI muscle of the 3rd and 4th interspaees were not significantly affected, however, changing from - 3.4~0 LR + 0.5 S E and - 2.6j0/o LR + 0.3 S E under control conditions to - 3.3~o L r + 0.6 SE and - 2.6}0 + 0.3 SE, respectively ( N S for each comparison). -6
/
-9
ExternallC Length (4th space) (%L R) Post Parasternal (1.6 spaces)
-6 Parasternal LC
Length (3rd space) (%L r', Post Denervation
0
c~o
g
o
-3
o°
o
ExternallC
(4th space) Denervation
*3 o
oo o
o
o oo --6
a
,-3
-3 0
,
'
i
-3
-;
-9
Parasternal IC Lengtl~ Control I%t R
-6
External IC Length (4th space) (%LR) Post Parasternal Denervation ~I-6 space'
Fig. 4. Identity plots relating active external intercostal shortening before (X-axis) and after (Y-axis) denervation in the 3rd (left hand panel) and 4th (right band panc]) interspaccs. Thc parasternal intercostal
muscles (Ist through 6th interspaccs) had been previously denervated. External intercostal denervation under these conditions usually resulted in their lengthening. See text for further description.
301
INTERCOSI-AI. MUSCLE ACTION
//
ExternallC Length (4th space; -3 (%Lr) Post ExternallC (3rd space) Denervation
/ i
0
1
-6 ExternalIC Length(4th space) Control (LR%) Fig. 5. Identity plot relating active external intercostal (El) shortening (4th interspace) before (X-axis) and -3
after (Y-axis) denervating the El muscle in the 3rd interspace. In each animal, El denervation (3rd interspace) resulted in an increase in El shortening in the 4th interspace.
Intact Para3ternallntercostal LMG 10 ~3rd space)(a.u.) 0E External Intercostal I:MG (3rd space) (a.u.)
lS I 0
/.~!
Post ParasternalIC Denervation ( 1-6 spaces)
Post ExternallC Denervation (3rd space)
/~
~ '3
/j ~ .
/ 3~X
' - - ~ ~
+1.5,Axial Displacement (4th rib) (mm)
0
t
.~7/Ak%//' -
~/k_~S'~ _~-
-1.5 Volume (ml)
300 [0L
_/Lr----q 2s
Fig. 6. Effect ofipsilateral parasternal and subsequent external intercostal (3rd space) denervation on the axial motion of the 4th rib in the mid-axillary line in one animal. The dashed line indicates the resting position of the 4th rib. Parasternal dcnervation resulted in an increase in EMG activity in the 3rd space. The 4th rib continued to move cephalad and above resting length during inspiration. Following external intercostal denervation, however, the 4th rib moved caudally during inspiration. See text for further explanation.
302
A.F. DIMARCO et al.
Changes in length resulting from subsequent EI denervation are shown for each animal in fig. 4. Mean El length of the 3rd and 4th interspaces changed from inspiratory shortening to inspiratory lengthening of + 0.8~,, Lg + 0.6 SE and + 1.2~o LR + 0.4 following their respective denervations (P < 0.01, for each comparison). The effects of El denervation were not significantly different from the effect of PA denervation on PA shortening. In 7 of the 9 animals, the El of the 3rd interspace was cut first and in each animal the degree of active inspiratory shortening of El in the 4th interspace increased (fig. 5) following this denervation. Mean shortening of E1 in the 4th interspace increased from 2.6~o LR + 0.3 SE to 3.6~o LR + 0.5 SE (P < 0.05) following denervation of El in the 3rd interspace and this occurred without any significant change in the level of activation of the El in the 4th interspace. The effect of intcrcostal denervation on the axial motion of the 4th rib is shown for a representative animal in fig. 6. Under control conditions, the 4th rib moved caudally during expiration below resting length, most likely secondary to expiratory muscle contraction. With expiratory muscle relaxation, the rib moved back to resting length and then further cephalad above resting length. Motion above resting length occurred synchronously with intercostal muscle electrical activation. As with intercostal length changes, therefore, rib motion also consisted of both active and passive components. Ipsilateral PA denervation resulted in an increase in El EMG with little effect on the axial motion of the 4th rib. Subsequent EI denervation produced a marked alteration in the pattern of rib movement such that the 4th rib moved caudally during inspiration (see fig. 6). In the 4 animals studied, mean cephalad motion of the 4th rib above resting length (i.e. active component) was 1.2mm + 0.1 SE under control conditions, 1.1 mm _+ 0.2 mm following ipsilateral PA denervation (NS) and - 0.1 mm + 0.1 mm SE after El denervation (P < 0.01 compared to pre-EI denervation and control). Mean in El EMG in these 4 animals increased to 205~o + 23 SE of control values (P < 0.05).
Discussion
The observation that the EI of the upper rib cage are electrically active and always shortened during inspiration suggests that these muscles play an active role in expanding the rib cage during resting breathing. But, these results alone do not exclude the possibility that El shortening and rib motion over the lateral rib cage are predominantly passive in nature [as suggested by other investigators (DeTroyer and Ninane, 1986)], occurring secondary to contraction of the PA or other rib cage muscles. The findings of the present study, however, which demonstrate that the El maintain the same level of active inspiratory shortening and that rib elevation is maintained despite ipsilateral PA denervation, indicates that coincident PA contraction is not a prerequisite for elevation of the ribs over the lateral portion of the rib cage during resting breathing. Furthermore, the fact that denervation of the El eliminated its inspiratory shortening
INTERCOSTAL MUSCLE ACTION
303
and cephalad rib motion, indicates that EI neural activation is a major factor responsible for El shortening and that El shortening is, in turn, responsible for elevation of the ribs over the laterai portion of the rib cage. Since PA shortening is also dependent predominantly on its own activation, our results suggest that regional expansion of the rib cage in different interspaces and even in different portions of the same interspace is highly dependent upon regional muscle activation. The upper rib cage, therefore, has multiple degrees of freedom and regional loss of muscle activation can lead to rib cage distortion.
INTERCOSTAL MUSCt.E INI'ERACTIONS External intercostal-parasternal. Although EI shortening was maintained essentially unchanged following ipsilateral parasternal denervation, this level of shortening required an increased level of motor drive to the EI muscles. Although the relationship between electrical activation and muscle shortening under these mechanical conditions is not known, it is reasonable to assume that the level of El shortening would have been less had El motor drive not increased. This indicates that parasternal contraction also contributes to El shortening during resting breathing. Consistent with this conclusion, are the findings of another study performed in our laboratory (DiMarco et al., 1990) which demonstrated persistent EI shortening in the 3rd interspace (although significantly less than control values) despite its denervation and the subsequent further reduction of El shortening following denervation of the PA in the 3rd interspace. Consistent with their anatomic orientation (Miller etal., 1964), this analysis would suggest that the PA and the El in the same interspace are arranged mechanically in parallel. Externalintercostal-externalintercostal. Contraction of the external intercostal muscle of the immediate cephalad interspace (by lifting the rib of muscle insertion) might be expected to oppose shortening of the interspace below by a direct mechanical effect. However, initial denervation of the El in the 3rd interspace resulted in an increase in shortening of the EI in the 4th space. Furthermore, this occurred without a significant increase in the level of activation of the E1 in the 4th space, suggesting a purely mechanical effect. Also, consistent with their anatomic arrangement, this observation suggests that the El muscles of the lateral rib cage are arranged mechanically in series. Parasternal-parasternal. Following denervation of the PA in the 3rd interspace, this muscle either lengthened or remained at resting length during inspiration. Denervation of the remaining ipsilateral PA muscles resulted in even further lengthening of the PA in the 3rd interspace, however, suggesting that the PA located in adjacent interspaces contribute to PA shortening in the 3rd interspace. This finding is consistent with those of DeTroyer etal. (1988) who demonstrated that the degree of lengthening of the denervated PA during spontaneous breathing was less than that which occurred during
304
A.F. DIMARCO et al.
isolated phrenic nerve stimulation for the same fall in pleural pressure. This indicates that other contracting rib cage muscles also contribute to PA shortening. Further studies by DeTroycr and Farkas (1989) demonstrated that the PA muscles located in adjacent spaces contribute directly to PA shortening in the 3rd space, leading these investigators to conclude that these muscles are arranged mechanically in parallel. The results of this study, therefore, support this conclusion.
POSSIBI.E MECtlANISM OF EXTERNAL INTERCOSTAl. RECRUITMENT End-tidal Pco2 and tidal volume were not significantly affected by ipsilateral PA denervation, suggesting that the increase in EI motor activity in response to PA denerration was not mediated by chemical factors. Excitation of EI activity may have occurred, however, via a neural reflex, possibly secondly to segmental intercostal autogenic facilitation (Corda et al., 1965). As discussed above, El shortening has one component attributable to its own neural activation but also a separate component attributable to PA contraction. It is possible that loss of the PA component to inspiratory shortening may have activated muscle spindles sensitive to small reductions in the degree of EI shortening. The resultant increased afferent discharge from muscle spindles may then have caused a reflex enhancement of efferent motor activity at a segmental level to preserve EI shortening. It should be noted that the increase in EI activation was just sufficient to maintain EI shortening at approximately the same level as that occurring under control conditions. This finding suggests that local rib cage reflexes exist to maintain the 'desired' length change and consequent expansion of the lateral rib cage.
COMPARISON TO PREVIOUS STUDIES The results of the present investigation are consistent with previous studies by DeTroyer and Farkas (1989) who found, in phrenicotomized animals, that the PA (3rd interspace) continued to shorten following bilateral denervation of the PA in the 2nd through 5th interspaces, even during occluded breaths. They surmised that external intercostal and levator costae muscle contraction during inspiration were probably responsible for PA shortening. In a more recent study, these same investigators demonstrated that following bilateral phrenicotomy and parasternal denervation, the levator costac and external intercostals maintained an inspiratory action on the rib cage (DeTroyer and Farkas, 1989). It should be emphasized, however, that phrenicotomy is known to result in increases in hypercapnic drive to the intercostal muscles (Ninane et al., 1989) and subsequent bilateral PA denervation results in even greater hypercapnic drive (DeTroyer and Farkas, 1989) and possibly severe hypoxemia, as well. This previous study, therefore, was conducted at significantly increased levels of chemical drive; indeed, El and lcvator costae E M G increased to greater than 1000 and 900 per cent of control values,
INTERCOSTAL MUSCI,E ACTION
305
respectively. Furthermore, it is not possible to ascertain the relative contribution of these different muscle groups to rib cage expansion. In contrast, the present study was performed in eucapneic animals and clearly demonstrates that EI contraction contributes significantly to rib elevation even under these conditions. The present studies also confirm previous reports that the parasternal muscles always shorten during inspiration and that PA denervation essentially eliminates PA shortening (DeTroyer et al., 1988). In fact, in 7 of the 9 animals examined in the present investigation, PA denervation (3rd interspace) resulted in PA lengthening during inspiration. This indicates that these muscles are true inspiratory agonists whose neural activation actively contributes to rib cage expansion during eupneic breathing. The finding that the PA (3rd interspace) shortened to a similar degree of approximately 3~o LR during resting breathing is consistent with a previous study from our laboratory in anesthetized dogs (DiMarco etaL, 1990), those of Decramer and DeTroyer (1984) who examined 39 different PA muscles in 10 animals (3-4 interspaces/animal) and Ninane et al. (1986). The level ofinspiratory EI shortening observed in the present study is less than that found by Decramer et al. (1986) who also measured EI shortening in the 3rd and 4th interspaces. Since the level of intercostal activation is very sensitive to anesthesia, it is possible that differences in anesthetic level may account for this discrepancy. The degree of cephalad movement of the 4th rib during inspiration observed in the present study (1.2 mm) is approximately half that reported in previous investigations (DeTroyer and Kelly, 1982) in which axial movement of the 6th or 7th rib was measured. This difference could be explained, in part, by the fact that rib motion was measured over different regions of the rib cage in these two studies. A more likely explanation, however, is the fact that only the active portion of rib motion was included in the measurements of the present study, whereas it is not apparent that the active and passive components were distinguished by DeTroyer and Kelly. Since approximately 50~o of cephalad rib displacement was passive in nature [most likely secondar3' to expiratory muscle relaxation (see fig. 6)], the sum of active and passive cephalad displacement during inspiration, is quite similar to this previous report. In summary, the El of the lateral rib cage are capable of causing rib elevation over the lateral rib cage without coincident parasternal contraction during eupneic breathing. The same level of EI shortening and degree of rib elevation is maintained however at the expense of an increase in neural drive to these muscles indicating that PA contraction contributes to EI shortening during resting breathing. Finally, expansion ot different areas of the rib cage is most dependent upon the electrical activation of the regional musculature. The rib cage therefore has a multiple degree of freedom and regional loss of intercostal muscle activation is likely to result in distortion of the rib cage.
306
A.F. DIMARCO et al.
References Budzinska, K., G. Supinski and A.F. DiMarco (1989). lnspiratory action of separate external and parasternal intercostal muscle contraction. J. Appl. Physiol. 67: 1395-1400. Corda, M., G. Eklund and C. von Euler (1965). External intercostal and phrenic alpha motor responses to changes in respiratory load. Acta Physiol. Stand. 63: 391-400. D'Angclo, E. (1982). Inspiratory muscle activity during rcbreathing in intact and vagotomizcd rabbits. Respir. Physiol. 47: 193-213. Decramer, M. and A. DeTroyer (1984). Respiratory changes in parasternal intercostal length. J. Appl. Physiol. 57: 1254-1260. Decramer, M., S. Kelly and A. DeTroyer (1986). Respiratory and postural changes in intercostal muscle length in supine dogs. J. Appl. Physiol. 60: 1686-1691. DeTroyer, A. and S. Kelly (1982). Chest wall mechanics in dogs with acute diaphragm paralysis. J. Appl. Physiol. 53: 373-379. DcTroyer, A., S. Kelly, P. T. Macklcm and W. A. Zinn (1985). Mechanics of the intercostal space and actions of external and internal intercostal muscles. J. Clin. Invest. 75: 850-857. DcTroyer, A. and V. Ninane (1986). Respiratory function of intercostal muscles in supine dog: an electromyographic study. J. Appl. Physiol. 60: 1692-1699. DeTroyer, A., G. A. Farkas and V. Ninane (1988). Mechanics of the parasternal intercostals during occluded breaths in dogs. J. Appl. Physiol. 64: 1546-1553. DeTroyer, A. and G.A. Farkas (1989). Passive shortening of canine parasternal intercostals during breathing. J. Appl. Physiol. 66: 1414-1420. DiMarco, A. F., J. R. Romaniuk and G. S. Supinski (1990). Mechanical action of the parasternal and external intercostal muscles during eupneic breathing. J. Appl. Physiol. (in press). Miller, M.E., G.C. Christensen and E. Evans (1964). Anatomy of the Dog. Philadelphia: W.B. Saunders. Newman, S.L., J. Road, F. Bellemare, J.P. Clozel, C.M. Lavigne and A. Grassino (1984). Respiratory muscle length measured by sonomicrometry. J. Appl. Physiol. 56: 753-764. Ninane, V., M. Decramer and A. DeTroyer (1986). Coupling between triangularis sterni and parasternals during breathing in dogs. J. Appl. Physiol. 61: 539-544. Ninane, V., G.A. Farkas, R. Baer and A. DeTroyer (1989). Mechanism of rib cage inspiratory muscle recruitment in diaphragmatic paralysis. Am. Rev. Respir. Dis. 139: 146-149.
295
RESP 01719
Action of the intercostal muscles on the rib cage A.F. DiMarco, J.R. Romaniuk* and G.S. Supinski Department Of Medicine. MetroHealth Medical Center and Case Western Reserve University, Cleveland, Ohio 44109, U.S.A. (Accepted 24 July 1990) Abstract. Recent studies suggest that the parasternal muscles (PA) are primarily responsible for rib cage expansion during eupneic breathing with a much lesser role played by the interosseous external intercostals (EI). The purpose of the present investigation was to assess the capacity of the El to expand the rib cage during spontaneous breathing in the absence of coincident ipsilateral PA activation. In 9 anesthetized dogs, we measured PA EMG and length in the 3rd interspace and El EMG and length in the 3rd and 4th interspaces. During resting breathing, each muscle was electrically active and shortened to a similar degree, approximately 3 ~o of resting length. Following ipsilateral PA denervation ( I st through 6th interspaces), the level of El shortening in the 3rd and 4th interspaces was maintained, but with an increase in neural drive to these muscles. The parasternal muscle in the 3rd interspace lengthened during inspiration. Subsequent sequential dencrvation of El in the 3rd and 4th interspaces resulted in their lengthening. In 4 additional animals, axial motion of the 4th rib was measured in the mid axillary line. Ipsilater',d PA denervation had no significant effect on rib motion. External intercostal denervation (3rd interspace), on the other hand, had a substantial impact on rib motion, causing the 4th rib to move in the caudal direction during inspiration. Our results indicate that: (a) the EI of the lateral rib cage are capable of elevating the ribs during inspiration independent of PA contraction; (b) PA contraction contributes to El shortening during eupneic breathing and (c) regional loss of muscle activation results in local rib cage distortion, suggesting that the upper rib cagc has multiple degrees of freedom.
Animal, dog; EMG, of respiratory muscles; Respiratory muscles, parasternal intercostals, interosseous external intercostals; Rib cage, movement and intercostal muscles
Both the parasternal (PA) and external interosseous intercostal (El) muscles of the upper rib cage are electrically active and shorten during inspiration (D'Angelo, 1982; Decramer and DeTroyer, 1984; DeTroyer and Ninane, 1986; Decramer et al., 1986). There is some controversy, however, concerning the relative importance of these two muscle groups in producing rib cage expansion. Some studies have suggested that the PA muscles are the primary agonists producing rib cage expansion with a much lesser Correspondence to: A.F. DiMarco, Metrollealth Medical Center, Pulmonary Division, 3395 Scranton Road, Cleveland, OH 44109, U.S.A. * Present address: Dept. of Ncurophysiology, Polish Academy of Sciences, Medical Research Center. Warsaw, Poland. 0034-5687/90:'S03.50
(~ 1990 Elsevier Science Publishers B.V. (Biomedical Division)
296
A.F. DIMARCO et al.
role played by the EI muscles. In fact, recent studies suggest that the interosseous intercostal muscles may subserve more important postural, as compared to ventilatory, functions (Decramer et al., 1986). This contention is based, in part, upon the findings that there are much greater changes in interosseous intercostal muscle length associated with rotation of the trunk compared to that associated with respiration. Other studies, however, have demonstrated that the El, like the PA muscles, are capable of producing large inspired volumes and changes in airway pressure (Budzinska et al., 1989), suggesting an important ventilatory role. These latter studies, however, were somewhat artificial in that they were performed utilizing electrical stimulation techniques and under conditions of near maximal intercostal muscle activation. The purpose of the present study, therefore, was to assess the capability of the external intercostal muscles to elevate the ribs during their spontaneous activation. It the external intercostals are, in fact, capable of elevating the ribs while contracting alone, their level of shortening, an index of rib elevation as well as axial motion of the ribs, should be maintained in the absence of parasternal contraction. External intercostal EMG and length and rib motion was assessed, therefore, both before and 'after denervating the ipsilateral PA musclcs in anesthetized dogs during resting breathing. To separate that portion of the El length changes and rib motion due to their own motor activation from that caused by other contracting inspiratory muscles, measurements were also made following subsequent EI denervation.
Methods
Experiments were performed in 13 supine dogs whose weight varied between 16 and 20 kg. Animals were anesthetized with pentobarbital sodium (Nembutal) at an initial dose of 25 mg/kg given intravenously. Supplemental doses (2-4 mg/kg) were given as needed to maintain a stable anesthetic level, i.e. absent response to noxious stimuli, but titrated to maintain an intact corneal reflex and normal end-tidal Pco2 throughout the experiment. A large bore endotracheal tube (ID = 9 mm) was sewn into the trachea in the mid-cervical region. An arterial catheter was placed in the femoral artery for blood pressure monitoring (Statham P23) and intermittent blood gas analysis. Another catheter was placed in the femoral vein to administer IV fluids and additional doses of the anesthetic agent. Changes in end-tidal Pco, were monitored at the tracheal opening via a rapid responding CO2 analyzer (Beckman LB-2). Body temperature was maintained at 38 + 0.5 °C by means of a homeothermic blanket (Harvard Apparatus, Cambridge, MA). Tidal volume was obtained by electrical integration of the flow signal from a pneumotachograph (Fleisch 1), connected to a pressure transducer (Validyne MP 45). Tracheal pressure was measured with a separate differential pressure transducer (Validyne MP 45). All signals were recorded on a 12-channel Electronics for Medicine oscilloscopic recorder. The superficial muscles of the rib cage including the pectorals, transvcrsus costarum and portions of the serratus anterior and scalenous muscles were sectioned unilaterally
INTERCOSTAL MUSCI.E ACTION
297
to facilitate access to the intercostal musculature. Bipolar stainless steel electrodes were then implanted under direct vision into the parasternal muscle (PA) of the 3rd intercostal space (approximately 2 cm lateral to the sternum) and into the external intercostal (El) muscles of the 3rd and 4th intercostal spaces (approximately 1 cm dorsal to the midaxillary line). Mass efferent activities of each muscle were amplified, rectified and processed by RC circuits containing third order low pass filters with a time constant of 100 msec (Charles Ward Enterprises, Ardmore, PA) to provide signals approximating moving averages of the recorded activity. Parasternal and El muscle length was measured in the same interspaces by sonomicrometry (Newman et al., 1984). Pairs of piezoelectric crystals (Model 120, Triton technology, San Diego, CA) were implanted into the PA muscle just lateral to the sternum and into the El near the mid-axillary line, just medial to the recording EMG electrodes. To facilitate placement of the piezoelectric crystals in the EI muscles, the animal was rotated to the opposite side and subsequently returned to the supine posture. Changes in intercostal muscle length were expressed as a percentage of resting length (~o LR). Similar to previously described methods (DeTroyer et al., 1985), a hook was screwed into the 4th rib in the mid-axillary line. An inelastic thread attached to the hook was passed over a pully connected to an appropriately oriented linear displacement transducer (Model 33-03-981, Brush Instruments, Cleveland, Ohio) to measure axial rib motion. Resting length of the intercostal muscles and rib position were determined by hyperventilation-induced apnea. In some animals, it was also necessary to administer small doses of succinylcholine to achieve complete apnea. Protocol. In 9 animals, baseline measurements of muscle shortening and peak EMG were made for each intercostal muscle and repeated after each of the following maneuvers. First, the internal intercostal nerve in the 3rd interspace was sectioned to denervate the parasternal muscle in this interspace. Subsequently, the internal intercostal nerves of the ipsilateral 1st, 2nd, 4th, 5th and 6th interspaces were sectioned just lateral to the costochondraljunctions to eliminate PA action over the ipsilateral rib cage. Denervation of the PA was limited to the ipsilateral rib cage since it was felt that these muscles have the greatest influence on the ipsilateral rib cage motion during resting breathing. Finally, the El nerves of the 3rd and 4th interspaces were sequentially sectioned. Following each muscle denervation, a 10-15 rain recovery period was allowed to elapse before measurements were repeated. Complete denervation of the PA (3rd interspace) and El muscles (3rd and 4th interspaces) was confirmed by the absence of inspiratory EMG activity in each interspace. An average of 3 breaths were taken under each condition to determine the level of E M G and muscle shortening. In 4 separate animals, axial motion of the 4th rib and El EMG were monitored under control conditions, following ipsilateral PA dcnervations (lst through 6th spaces) and subsequent El (3rd space) denervation. An average of 3 breaths was taken to determine the level of EMG and rib motion. The effect of denervation of each muscle was asscssed statistically by paired t-tests.
298
A.F. DIMARCO et al.
Statistical comparison between the 3 muscles were made by employing a repeated measures analysis of variance and p o s t hoc t-test.
Results
Mean inspired volumc and end-tidal Pco~ were 350 ml _+ 23 SE and 40 mmHg + 1 SE, respectively, under control conditions and were not significantly affected by ipsilateral PA and EI denervation, lnspiratory electrical activity and active inspiratory shortening was present in each of the 27 intercostal muscles studied. The effects of initial denervation of the parasternal muscles followed by El muscle denervation is shown for a representative animal in fig. 1. The PA of the 3rd interspace and EI of the 3rd and 4th interspaces were electrically active and shortened during inspiration. An initial rapid shortening to resting length is followed by more gradual shortening below resting length. The initial rapid phase of shortening has been described previously for the PA and is attributable to expiratory muscle relaxation (Ninane et al., 1986). The subsequent shortening below resting length during inspiration is the subject of this study and is referred to as active shortening. This phase of shortening occurs in synchrony with the onset of intercost',d EMG activity. Denervation of PA of the 3rd space resulted in the elimination of active PA shortening during inspiration. As shown in this example, PA denervation had little effect on the level of EI activation or degree of inspiratory shortening of the EI in the 3rd interspace. There was a small increase in EI shortening in the 4th interspace with essentially no change in its EMG. Following subsequent denervation of the other ipsilateral PA muscles (lst, 2nd, 4th, 5th and 6th interspaces), EI EMG of both the 3rd and 4th interspaces increased. This increased EMG activity, however, was not associated with an appreciable change in peak inspiratory EI shortening. Subsequent denervation of EI muscle of the 3rd interspace resulted in inspiratory lengthening of this muscle and a small increase in inspiratory shortening of El in the 4th interspace without change in the level of its motor activation. Subsequent denervation of EI in the 4th space resulted in inspiratory lengthening of this muscle as well. The level of active inspiratory shortening under control conditions for the PA (3rd interspace) and El (3rd and 4th interspaces)was 3.3 ~o LR +_ 0.6 S E, 3.4~o LR +_ 0.5 S E and 2.6~, LR + 0.3 SE, respectively, and not significantly different from each other. The effects of initial PA denervation (3rd interspace) on PA shortening is shown for each animal in fig. 2. In each animal, there was a marked reduction in active PA shortening. Mean PA length (3rd interspace) changed from an inspiratory shortening under control conditions to + 0.7% LR +_ 0.2 SE following PA denervation. PA denervation resulted in a small increase in EMG to the El in the 3rd and 4th interspaces (NS), as shown in fig. 3 and no significant change in El length. Denervation of the PA (lst, 2nd and 4-6th interspaces) resulted in further lengthening of the PA (3rd interspace) to + 1.1 ~o LR + 1.2 SE (NS). Mean El EMG of the 3rd and 4th interspaces increased further following ipsilateral PA denervation (lst through 6th interspaceXP < 0.01 compared to
299
INTERCOSTAL MUSCLE ACTION PoM Palaste~r~| IC D e r ~ f v a l i o n (3rd space)
Control
P,last¢Inal
Po~t P a l a s l e r ~ l
Posl Exter~,al
~ n , ervalion
IC [~,,efvatioqrl (3rd ~ ( . e )
(1-6 spaces)
Pos! I xternal IC [ ~ e r v a t l o n (41h space)
0
(a,uP
Ixt~nal K:
lao)
lo E
o
~ - r ~ - ~ , - ~
length I3ld sp~e) (%tR)
-s
I%I~)
.(,
~
--"--"'--~-- ~
-
- ~=
L
\
[ ,le,.JI IC Ir:,glh t41t, ~l,~e p I%IR )
4
' ,
.-
/
~
~\ '
:\/
'k./
.S'
Fig. 1. Effect of sequential denervation of the parasternal (PA) and external intercostal (El) muscles in a single animal. Initial PA denervation in the 3rd interspace eliminated active shortening of that muscle. Subsequent denervation of the other ipsilatcral PA (lst through 6th interspaces) resulted in an increase in El muscle activation but no significant changes in active El shortening. Subsequent El denervation in the 3rd and 4th intcrspaces resulted in El lengthening during inspiration. See text for further description.
-9
-6 Parasternal IC
length (3rd space)
-3
(%LR)
Post Denervation 0
0
0 00
O0 0 0
+3 0
I
1
I
-3
-6
-9
Parasternal IC Length Control (%LR) Fig. 2. Identity' plot relating active parasternal intercostal shortening in the 3rd interspace before (X-axis) and after (Y-axis) its dencrvation. Dencrvation of this muscle virtually eliminated its inspiratory shortening.
300
A.I.. DIMARCO 200
3rd space
et al.
4th space
100 x
Control F77";I Post Parasterna: Denervation (3rd space) Post Paras:err'al Denervation (1-6 space) t"ig. 3. Effect of parasternal (PA) denervation on peak external intercostal (El) E M G in the 3rd and 4th
interspaces. Denervation of the PA muscle in the 3rd interspace resulted in a small increase in El E M G (NS). Subsequcnt dcnervation of thc ipsilateral PA intercostals in the 1st through 6th spaces resulted in significant incrcases in pcak EI EMG in both the 3rd and 4th interspaces.
control values). The mean level of active inspiratory shortening o f the EI muscle of the 3rd and 4th interspaees were not significantly affected, however, changing from - 3.4~0 LR + 0.5 S E and - 2.6j0/o LR + 0.3 S E under control conditions to - 3.3~o L r + 0.6 SE and - 2.6}0 + 0.3 SE, respectively ( N S for each comparison). -6
/
-9
ExternallC Length (4th space) (%L R) Post Parasternal (1.6 spaces)
-6 Parasternal LC
Length (3rd space) (%L r', Post Denervation
0
c~o
g
o
-3
o°
o
ExternallC
(4th space) Denervation
*3 o
oo o
o
o oo --6
a
,-3
-3 0
,
'
i
-3
-;
-9
Parasternal IC Lengtl~ Control I%t R
-6
External IC Length (4th space) (%LR) Post Parasternal Denervation ~I-6 space'
Fig. 4. Identity plots relating active external intercostal shortening before (X-axis) and after (Y-axis) denervation in the 3rd (left hand panel) and 4th (right band panc]) interspaccs. Thc parasternal intercostal
muscles (Ist through 6th interspaccs) had been previously denervated. External intercostal denervation under these conditions usually resulted in their lengthening. See text for further description.
301
INTERCOSI-AI. MUSCLE ACTION
//
ExternallC Length (4th space; -3 (%Lr) Post ExternallC (3rd space) Denervation
/ i
0
1
-6 ExternalIC Length(4th space) Control (LR%) Fig. 5. Identity plot relating active external intercostal (El) shortening (4th interspace) before (X-axis) and -3
after (Y-axis) denervating the El muscle in the 3rd interspace. In each animal, El denervation (3rd interspace) resulted in an increase in El shortening in the 4th interspace.
Intact Para3ternallntercostal LMG 10 ~3rd space)(a.u.) 0E External Intercostal I:MG (3rd space) (a.u.)
lS I 0
/.~!
Post ParasternalIC Denervation ( 1-6 spaces)
Post ExternallC Denervation (3rd space)
/~
~ '3
/j ~ .
/ 3~X
' - - ~ ~
+1.5,Axial Displacement (4th rib) (mm)
0
t
.~7/Ak%//' -
~/k_~S'~ _~-
-1.5 Volume (ml)
300 [0L
_/Lr----q 2s
Fig. 6. Effect ofipsilateral parasternal and subsequent external intercostal (3rd space) denervation on the axial motion of the 4th rib in the mid-axillary line in one animal. The dashed line indicates the resting position of the 4th rib. Parasternal dcnervation resulted in an increase in EMG activity in the 3rd space. The 4th rib continued to move cephalad and above resting length during inspiration. Following external intercostal denervation, however, the 4th rib moved caudally during inspiration. See text for further explanation.
302
A.F. DIMARCO et al.
Changes in length resulting from subsequent EI denervation are shown for each animal in fig. 4. Mean El length of the 3rd and 4th interspaces changed from inspiratory shortening to inspiratory lengthening of + 0.8~,, Lg + 0.6 SE and + 1.2~o LR + 0.4 following their respective denervations (P < 0.01, for each comparison). The effects of El denervation were not significantly different from the effect of PA denervation on PA shortening. In 7 of the 9 animals, the El of the 3rd interspace was cut first and in each animal the degree of active inspiratory shortening of El in the 4th interspace increased (fig. 5) following this denervation. Mean shortening of E1 in the 4th interspace increased from 2.6~o LR + 0.3 SE to 3.6~o LR + 0.5 SE (P < 0.05) following denervation of El in the 3rd interspace and this occurred without any significant change in the level of activation of the El in the 4th interspace. The effect of intcrcostal denervation on the axial motion of the 4th rib is shown for a representative animal in fig. 6. Under control conditions, the 4th rib moved caudally during expiration below resting length, most likely secondary to expiratory muscle contraction. With expiratory muscle relaxation, the rib moved back to resting length and then further cephalad above resting length. Motion above resting length occurred synchronously with intercostal muscle electrical activation. As with intercostal length changes, therefore, rib motion also consisted of both active and passive components. Ipsilateral PA denervation resulted in an increase in El EMG with little effect on the axial motion of the 4th rib. Subsequent EI denervation produced a marked alteration in the pattern of rib movement such that the 4th rib moved caudally during inspiration (see fig. 6). In the 4 animals studied, mean cephalad motion of the 4th rib above resting length (i.e. active component) was 1.2mm + 0.1 SE under control conditions, 1.1 mm _+ 0.2 mm following ipsilateral PA denervation (NS) and - 0.1 mm + 0.1 mm SE after El denervation (P < 0.01 compared to pre-EI denervation and control). Mean in El EMG in these 4 animals increased to 205~o + 23 SE of control values (P < 0.05).
Discussion
The observation that the EI of the upper rib cage are electrically active and always shortened during inspiration suggests that these muscles play an active role in expanding the rib cage during resting breathing. But, these results alone do not exclude the possibility that El shortening and rib motion over the lateral rib cage are predominantly passive in nature [as suggested by other investigators (DeTroyer and Ninane, 1986)], occurring secondary to contraction of the PA or other rib cage muscles. The findings of the present study, however, which demonstrate that the El maintain the same level of active inspiratory shortening and that rib elevation is maintained despite ipsilateral PA denervation, indicates that coincident PA contraction is not a prerequisite for elevation of the ribs over the lateral portion of the rib cage during resting breathing. Furthermore, the fact that denervation of the El eliminated its inspiratory shortening
INTERCOSTAL MUSCLE ACTION
303
and cephalad rib motion, indicates that EI neural activation is a major factor responsible for El shortening and that El shortening is, in turn, responsible for elevation of the ribs over the laterai portion of the rib cage. Since PA shortening is also dependent predominantly on its own activation, our results suggest that regional expansion of the rib cage in different interspaces and even in different portions of the same interspace is highly dependent upon regional muscle activation. The upper rib cage, therefore, has multiple degrees of freedom and regional loss of muscle activation can lead to rib cage distortion.
INTERCOSTAL MUSCt.E INI'ERACTIONS External intercostal-parasternal. Although EI shortening was maintained essentially unchanged following ipsilateral parasternal denervation, this level of shortening required an increased level of motor drive to the EI muscles. Although the relationship between electrical activation and muscle shortening under these mechanical conditions is not known, it is reasonable to assume that the level of El shortening would have been less had El motor drive not increased. This indicates that parasternal contraction also contributes to El shortening during resting breathing. Consistent with this conclusion, are the findings of another study performed in our laboratory (DiMarco et al., 1990) which demonstrated persistent EI shortening in the 3rd interspace (although significantly less than control values) despite its denervation and the subsequent further reduction of El shortening following denervation of the PA in the 3rd interspace. Consistent with their anatomic orientation (Miller etal., 1964), this analysis would suggest that the PA and the El in the same interspace are arranged mechanically in parallel. Externalintercostal-externalintercostal. Contraction of the external intercostal muscle of the immediate cephalad interspace (by lifting the rib of muscle insertion) might be expected to oppose shortening of the interspace below by a direct mechanical effect. However, initial denervation of the El in the 3rd interspace resulted in an increase in shortening of the EI in the 4th space. Furthermore, this occurred without a significant increase in the level of activation of the E1 in the 4th space, suggesting a purely mechanical effect. Also, consistent with their anatomic arrangement, this observation suggests that the El muscles of the lateral rib cage are arranged mechanically in series. Parasternal-parasternal. Following denervation of the PA in the 3rd interspace, this muscle either lengthened or remained at resting length during inspiration. Denervation of the remaining ipsilateral PA muscles resulted in even further lengthening of the PA in the 3rd interspace, however, suggesting that the PA located in adjacent interspaces contribute to PA shortening in the 3rd interspace. This finding is consistent with those of DeTroyer etal. (1988) who demonstrated that the degree of lengthening of the denervated PA during spontaneous breathing was less than that which occurred during
304
A.F. DIMARCO et al.
isolated phrenic nerve stimulation for the same fall in pleural pressure. This indicates that other contracting rib cage muscles also contribute to PA shortening. Further studies by DeTroycr and Farkas (1989) demonstrated that the PA muscles located in adjacent spaces contribute directly to PA shortening in the 3rd space, leading these investigators to conclude that these muscles are arranged mechanically in parallel. The results of this study, therefore, support this conclusion.
POSSIBI.E MECtlANISM OF EXTERNAL INTERCOSTAl. RECRUITMENT End-tidal Pco2 and tidal volume were not significantly affected by ipsilateral PA denervation, suggesting that the increase in EI motor activity in response to PA denerration was not mediated by chemical factors. Excitation of EI activity may have occurred, however, via a neural reflex, possibly secondly to segmental intercostal autogenic facilitation (Corda et al., 1965). As discussed above, El shortening has one component attributable to its own neural activation but also a separate component attributable to PA contraction. It is possible that loss of the PA component to inspiratory shortening may have activated muscle spindles sensitive to small reductions in the degree of EI shortening. The resultant increased afferent discharge from muscle spindles may then have caused a reflex enhancement of efferent motor activity at a segmental level to preserve EI shortening. It should be noted that the increase in EI activation was just sufficient to maintain EI shortening at approximately the same level as that occurring under control conditions. This finding suggests that local rib cage reflexes exist to maintain the 'desired' length change and consequent expansion of the lateral rib cage.
COMPARISON TO PREVIOUS STUDIES The results of the present investigation are consistent with previous studies by DeTroyer and Farkas (1989) who found, in phrenicotomized animals, that the PA (3rd interspace) continued to shorten following bilateral denervation of the PA in the 2nd through 5th interspaces, even during occluded breaths. They surmised that external intercostal and levator costae muscle contraction during inspiration were probably responsible for PA shortening. In a more recent study, these same investigators demonstrated that following bilateral phrenicotomy and parasternal denervation, the levator costac and external intercostals maintained an inspiratory action on the rib cage (DeTroyer and Farkas, 1989). It should be emphasized, however, that phrenicotomy is known to result in increases in hypercapnic drive to the intercostal muscles (Ninane et al., 1989) and subsequent bilateral PA denervation results in even greater hypercapnic drive (DeTroyer and Farkas, 1989) and possibly severe hypoxemia, as well. This previous study, therefore, was conducted at significantly increased levels of chemical drive; indeed, El and lcvator costae E M G increased to greater than 1000 and 900 per cent of control values,
INTERCOSTAL MUSCI,E ACTION
305
respectively. Furthermore, it is not possible to ascertain the relative contribution of these different muscle groups to rib cage expansion. In contrast, the present study was performed in eucapneic animals and clearly demonstrates that EI contraction contributes significantly to rib elevation even under these conditions. The present studies also confirm previous reports that the parasternal muscles always shorten during inspiration and that PA denervation essentially eliminates PA shortening (DeTroyer et al., 1988). In fact, in 7 of the 9 animals examined in the present investigation, PA denervation (3rd interspace) resulted in PA lengthening during inspiration. This indicates that these muscles are true inspiratory agonists whose neural activation actively contributes to rib cage expansion during eupneic breathing. The finding that the PA (3rd interspace) shortened to a similar degree of approximately 3~o LR during resting breathing is consistent with a previous study from our laboratory in anesthetized dogs (DiMarco etaL, 1990), those of Decramer and DeTroyer (1984) who examined 39 different PA muscles in 10 animals (3-4 interspaces/animal) and Ninane et al. (1986). The level ofinspiratory EI shortening observed in the present study is less than that found by Decramer et al. (1986) who also measured EI shortening in the 3rd and 4th interspaces. Since the level of intercostal activation is very sensitive to anesthesia, it is possible that differences in anesthetic level may account for this discrepancy. The degree of cephalad movement of the 4th rib during inspiration observed in the present study (1.2 mm) is approximately half that reported in previous investigations (DeTroyer and Kelly, 1982) in which axial movement of the 6th or 7th rib was measured. This difference could be explained, in part, by the fact that rib motion was measured over different regions of the rib cage in these two studies. A more likely explanation, however, is the fact that only the active portion of rib motion was included in the measurements of the present study, whereas it is not apparent that the active and passive components were distinguished by DeTroyer and Kelly. Since approximately 50~o of cephalad rib displacement was passive in nature [most likely secondar3' to expiratory muscle relaxation (see fig. 6)], the sum of active and passive cephalad displacement during inspiration, is quite similar to this previous report. In summary, the El of the lateral rib cage are capable of causing rib elevation over the lateral rib cage without coincident parasternal contraction during eupneic breathing. The same level of EI shortening and degree of rib elevation is maintained however at the expense of an increase in neural drive to these muscles indicating that PA contraction contributes to EI shortening during resting breathing. Finally, expansion ot different areas of the rib cage is most dependent upon the electrical activation of the regional musculature. The rib cage therefore has a multiple degree of freedom and regional loss of intercostal muscle activation is likely to result in distortion of the rib cage.
306
A.F. DIMARCO et al.
References Budzinska, K., G. Supinski and A.F. DiMarco (1989). lnspiratory action of separate external and parasternal intercostal muscle contraction. J. Appl. Physiol. 67: 1395-1400. Corda, M., G. Eklund and C. von Euler (1965). External intercostal and phrenic alpha motor responses to changes in respiratory load. Acta Physiol. Stand. 63: 391-400. D'Angclo, E. (1982). Inspiratory muscle activity during rcbreathing in intact and vagotomizcd rabbits. Respir. Physiol. 47: 193-213. Decramer, M. and A. DeTroyer (1984). Respiratory changes in parasternal intercostal length. J. Appl. Physiol. 57: 1254-1260. Decramer, M., S. Kelly and A. DeTroyer (1986). Respiratory and postural changes in intercostal muscle length in supine dogs. J. Appl. Physiol. 60: 1686-1691. DeTroyer, A. and S. Kelly (1982). Chest wall mechanics in dogs with acute diaphragm paralysis. J. Appl. Physiol. 53: 373-379. DcTroyer, A., S. Kelly, P. T. Macklcm and W. A. Zinn (1985). Mechanics of the intercostal space and actions of external and internal intercostal muscles. J. Clin. Invest. 75: 850-857. DcTroyer, A. and V. Ninane (1986). Respiratory function of intercostal muscles in supine dog: an electromyographic study. J. Appl. Physiol. 60: 1692-1699. DeTroyer, A., G. A. Farkas and V. Ninane (1988). Mechanics of the parasternal intercostals during occluded breaths in dogs. J. Appl. Physiol. 64: 1546-1553. DeTroyer, A. and G.A. Farkas (1989). Passive shortening of canine parasternal intercostals during breathing. J. Appl. Physiol. 66: 1414-1420. DiMarco, A. F., J. R. Romaniuk and G. S. Supinski (1990). Mechanical action of the parasternal and external intercostal muscles during eupneic breathing. J. Appl. Physiol. (in press). Miller, M.E., G.C. Christensen and E. Evans (1964). Anatomy of the Dog. Philadelphia: W.B. Saunders. Newman, S.L., J. Road, F. Bellemare, J.P. Clozel, C.M. Lavigne and A. Grassino (1984). Respiratory muscle length measured by sonomicrometry. J. Appl. Physiol. 56: 753-764. Ninane, V., M. Decramer and A. DeTroyer (1986). Coupling between triangularis sterni and parasternals during breathing in dogs. J. Appl. Physiol. 61: 539-544. Ninane, V., G.A. Farkas, R. Baer and A. DeTroyer (1989). Mechanism of rib cage inspiratory muscle recruitment in diaphragmatic paralysis. Am. Rev. Respir. Dis. 139: 146-149.
