Ventilation and Gas Exchange during Exercise in Sickle Cell Anemia 1- 3 PAUL PIANOSI,4 S. J. A. D'SOUZA,5 DIXIE W. ESSELTINE, T. D. CHARGE, and ALLAN L. COATES
Introduction
Sickle cell anemia (SCA) is characterized by the propensity for HbS to polymerize in a hypoxic milieu, resulting in changes in the shape and deformability of the red blood cell. In physiologicterms, occlusion of the pulmonary capillaries would result in a region of ventilationl perfusion fY/Q) imbalance, not unlike that seen in pulmonary embolism, leading to increased dead space. This could be a permanent vascular occlusion leading to infarction or a functional occlusion where red cell movement is slowed but not completely arrested. Older patients, especially those who have experienced multiple episodes of "acute chest syndrome," are presumably more at risk for permanent impairment than are younger patients. Elevation of dead space has been found in adults with SCA during exercise (1) but not at rest. In order to achieve the level of alveolar ventilation required to maintain normocapnia in the face of increased dead space, overall ventilation is increased (1). However, there are other recognized physiologic derangements in sickle cell disease that might account for the hyperventilation seen during exercise. Miller and coworkers (1) concluded that in addition to greater dead space ventilation, an increased need for anaerobic metabolism in the more anemic patients and the hypoxemia seen in this disorder (2) contributed to the exaggerated ventilatory response to exercise. In children with SCA, Wall and coworkers (3) found mild hypoxemia but spirometric indices similar to those in control subjects. Alpert and associates (4) have demonstrated low exercisetolerance, and studied cardiovascular responses to exercise in these children, but there have been no studies on the ventilatory response to exercisein this group. The purpose of this investigation was to measure the ventilatory response to exercise and to assess the roles of increased physiologic dead space, anaerobic metabolism, and hypoxemia in determining this response in a group of children with SCA that included those who had suffered episodes of "acute chest syn226
SUMMARY Adults with sickle cell anemia (SCA) have restrictive lung Impairment, Increased alveolar dead space, and hypoxemia. These factors, together with Increased anaerobic metabolism, are thought to cause exercise hyperventilation. To assess the role of each of these In children, 34 patients with SCA and 16 control subjects performed pulmonary function and exercise tests. 1Wentyeight patients with SCA had spirometric values and lung volumes, and all but two patients with SCA had arterial saturation> 91% during exercise. Despite a low Vo 2 max (30.07 ± 6.55 ml/mln/kg), the ventilatory anaerobic threshold (VAT) In the patients occurred at a similar % Vo 2max as In the control subjects (69 ± 9% versus 63 ± 12%). The slope of the A VEl AVC0 2 relationship for sub-VAT work was steeper In the patients (29.4 ± 6.5 versus 24.7 ± 5.2, P = 0.01),and the ventilatory equivalent for CO2 (VElVC0 2) In steady-state exercise was greater In the patients than In the control sub0.03). End-tidal PC0 2 did not differ (38.3 ± 3.0 versus Jects (33.2 ± 3.5 versus 30.8 ± 3.5, P 39.2 ± 3.1), Indicating equivalent alveolar ventilation. The patients had a higher dead space:tldal volume ratio (VDNT) than did the control subjects (0.204 ± 0.033 versus 0.173 ± 0.024, P = 0.0005). The PaC0 2 was significantly lower In those with lower Hb, but there was no difference In pH. In conclusion, children with SCA have an Increased exercise ventilatory response caused In part by Increased physiologic dead space, and In part by their low Hb. The greater dead space may be the result of sickle cells Impairing capillary perfusion to ventilated alveoli.
=
AM REV RESPIR DIS 1991; 143:228-230
drome" and those who had experienced a more benign course. Methods Study Population Subjects were recruited from the Sickle Cell Anemia Clinic at the Montreal Children's Hospital. Inclusion criteria were: age, 6 to 19 yr; height> 120 em (lower height limit of the exercise cycle ergometer), HbSS or HbSC on hemoglobin electrophoresis, absence of congenital heart disease or hypertension, and not participating in a regular transfusion program. Patients were studied a minimum of 4 wk after any illness, and 3 months after any hospitalization. All patients were studied in their normal stable condition as judged by hemoglobin level and clinical assessment. Control subjects were recruited from the families of hospital personnel or from siblings of the patients with SeA (HbAA), and they were free of cardiopulmonary disease, The hemoglobin level and type were known in all siblings, and screening was offered to nonrelated control subjects. The study received approval from the Institutional Review Board, and all subjects gave informed, written consent. The charts were reviewed by one author (SJAD'S), and those responsible for the exercise testing had no previous knowledge of the results. Patients who exhibited a more severe clinical course were compared with thei~ counterparts with milder disease. A more severeclinicalcourse wasdefined as having been
hospitalized, or having received an exchange transfusion, for treatment of one or more episodes of "acute chest syndrome," or having been admitted to the ICU for any reason related to management of a vaso-occlusive crisis. It was hypothesized that such subjects would be at highest risk of developing irreversible changes in their pulmonary vasculature because of the sickling process, which would be made manifest by an elevated VD/VT ratio.
Protocol All testing was done the same day (morning or afternoon) after a light meal. Subjects had electrocardiograms to look for possible arrhythmias and then performed pulmonary function testing consisting 0 f spirometry and
(Received in originalform September 22, 1989and in revised form May 18, 1990) 1 From the Respiratory Medicine and Hematology Services, McGill University-Montreal Children's Hospital Research Institute, Montreal, Quebec, Canada. 2 Publication no. 90-023 of the McGill University-Montreal Children's Hospital Research Institute. 3 Correspondence and requests for reprints should be addressed to Dr. Allan L. Coates, The Montreal Children's Hospital, 2300 Tupper Street, Montreal, Quebec H3H IP3, Canada. 4 Fellow of the Canadian Cystic Fibrosis Foundation. S Funded by Le Fonds de La Recherche en Sante du Quebec.
VENTILATION AND GAS EXCHANGE DURING EXERCISE IN SICKLE CELL ANEMIA
plethysmographic lung volume measurement. They then performed a progressive exercise test (Jones Stage I) (5) on an electronically braked cycle ergometer (Quinton Uniwork Model 884 for those between 120and 150cm; Lode Corival 300 for those> 150 em) to exhaustion. The final load was taken as their maximal work load (Wmax). The increment in work load each minute varied according to the size of the subject, being 8 W for the smallest children and as much as 20 W for adolescents, such that the test would be expected to last from 5 to 10 min. After a l-h rest period, during which subjects usually had a snack, they pedaled at 500/0 of their Wmax at a frequency of 50 to 70 rpm for 5 to 7 min (Jones Stage III). The work load of 500/0 Wmax was chosen because past experience had shown that not all untrained or sick subjects were capable of maintaining a work load of two-thirds Wmax, the usual standard load, for the time necessary to achieve good steadystate values. Steady-state conditions were considered present when fluctuations of not more than ± 0.10/0 in mixed expired gas concentrations and ± 5 bpm in heart rate lasting at least 30 s were noted after the third minute of exercise (5). At this time, arterialized capillary blood specimens consisting of two or three 125-J.11 tubes were obtained. In two subjects the samples wererejected because of poor blood flow from the finger-prick. No blood sampling was done on control subjects during exercise. During exercise, subjects wore noseclips and breathed through a two-way, nonrebreathing valve(Hans Rudolph, Kansas City, MO or Koegel, Houston, TX). Apparatus dead space varied with subject size but was generally set up as follows: 55 ml for subjects < 140 em, 65 ml for subjects 140 to 160 em, and 75 ml for subjects> 160em, Although the resistance of the valves also varied, being greater with the smaller apparatus, size-related variability in flow rates ensured that resistance remained negligible.
Measurements Heart rate (HR) and saturation were monitored using a pulse oximeter (Model N-2oo; Nellcor, Hayward, CA). Inspired ventilation was measured with a dry gas meter (Parkinson-Cowan, Manchester, UK). Expired gas was sampled from a mixing chamber with a fan, the volume of which was adjusted to approximately 750/0 of the subject's vital capacity. The expirate passed through 10to 20 mesh CaS04 (Drierite) to remove moisture, and the concentrations of O 2 and CO 2 (FE02, FEc0 2) were measured by rapidly responding gas analyzers (Model SA-3; Applied Electrochemistry, Pasadena, CA, and Model LB-2; Beckman Instruments, Fullerton, CA). Both were calibrated with room air and a test gas of known concentration (gravimetric method) before each stage of exercise testing. End-tidal CO 2 concentration was also sampled via a heated line at the mouthpiece over 3 to 6 breaths during the latter half of each minute, Output signals for HR, VI, FE02, FEC02' and FETC02 wererecorded on an eight-
channel chart recorder (Model 7758A; Hewlett-Packard, Waltham, MA), whereas the numerical value for saturation was recorded manually three to five times per minute. From the inspired minute ventilation (VI), FE02, and FEc0 2, measured over the final 15 to 20 s of each minute, expired minute ventilation (VE), oxygen consumption (V02), and carbon dioxide production (VC02) were calculated by standard methods (5) and considered as the values representative at that work load. Valuesfor VE werecorrected to BTPS conditions, whereas V02and VC02 were expressed as STPn values. The V02for the last work load was taken as the subject's V02max. The ventilatory anaerobic threshold (VAT) was determined by the V-slope method (6), and it was described both in terms of absolute V02 (ml/min/kg) and as % V02max. Blood specimens were stored on ice until analyzed for pH, P0 2, and Pc0 2 (ABL3 blood gas analyzer; Radiometer, Copenhagen, Denmark) within 20 min of sampling.
Analysis A test was considered acceptable if there was a progressive increase in VE, V02, and VC02. Transient increases in ventilation were felt to represent hyperventilation, and tests with those types of changes were rejected. The ventilatory response to exercise was assessed in two ways. First, the slope of the change in VE versus the change in VC02 (AVElAVC02) was calculated over the sub-VAT range of incremental exercise, and then ventilatory equivalent for CO 2during steady-state exercise(VEIVc02) was calculated. Dead space (Vo) was calculated using the Bohr equation using Pe02values obtained in steady-state exercise, with the appropriate correction made for apparatus dead space. Because Vn increases with increasing tidal volume (VT), and to facilitate comparison between different-sized subjects, the ratio of dead space to tidal volume (Vn/VT) was examined. As blood sampling was not done on control subjects, the PETC02 was used as an estimate of Pac02 in the calculation of Vn/VT in this group. Thus, when the groups werecompared, the calculated value for both control subjects and patients with SCA was based on PETC02. This was not unreasonable since the difference between simultaneously obtained PETC02 and Pac02 in the patients with SCA was 2.25 ± 2.27 mm Hg (range, +5.6to -2.7). Grant and coworkers (7) cite similar differences that wereinsignificant, and the use of PETC02' even in the presence of lung disease, appears to be more reliable than previously thought (7). The relation between the VE/VC02 and VD/VT in patients with SCA during steady-state exercise was explored by linear regression techniques, and in this case the values for Vn/VT were those based on Pac02. Comparisons between patients and control subjects were tested for statistical significance using the two-tailed unpaired t test, with a < 0.05. Statistical analysis was done using a Minitab PC program (8).
227
Results
A total of 50 successful studies were accomplished. There were three patients with SCA and three control subjects for whom either hyperventilation, lack of effort, or technical problems were grounds for excluding the tests from further analysis. Of the 34 patients with SCA, 18 had HbSS, with levels ranging from 65 to 106 giL and 16had HbSC, with levelsranging from 109 to 133 giL. This represented 80070 of patients in the clinic who met the inclusion criteria. Of those studied, 25070 were considered to have severe clinical disease; the same proportion among those not studied was 20070. Sixteen control subjects were studied, of whom nine were documented to have HbAA; the remaining seven declined a screening test. Demographic data are displayed in table 1, which shows that patients with SCA and control subjects were very closely matched for sex, age, height, and weight. The main difference between the groups was hemoglobin levels. Pulmonary function data are listed in table 2. An in-depth report of these will appear in a separate article, but it is apparent from the table that patients with SCA had lower spirometric indices than did control subjects. Twenty-eight of the patients had spirometric indices within normal limits (9), whereas all control subjects were within the accepted normal range. All patients with SCA completed the protocol without incident. All but one gave what appeared to be maximal performance as judged by effort and peak HR ~ 174. One subject had a peak HR of 156, but his effort was judged to be maximal nonetheless, whereas another achieved a peak HR of 172, but her effort appeared submaximal, and therefore her Vo2 max data were excluded from analysis.Ventilation at maximal exercisewas greater in control subjects than in patients with SCA (1.396 ± 0.285 versus 1.204 ± 0.287Llminlkg, p = 0.03), but there was
TABLE 1 PHYSICAL CHARACTERISTICS OF PATIENTS WITH SCA AND CONTROL SUBJECTS· Patients M/F Age, yr Height, cm Weight, kg Hb, gIL
17/17
13.0 153 44 101
± 3.8 ± 15 ± 15 ± 18
Control Subjects
8/8 12.3 ± 3.5 152 ± 15 46 ± 16
Definitionof abbreviations: SeA = sickle cell anemia; Hb = hemoglobin. * Values are mean ± SO.
228
PIANOSI, D'SOUZA, ESSELTINE, CHARGE, AND COA'!'ES
TABLE 2 PULMONARY FUNCTION* AND AEROBIC EXERCISE PARAMETERS OF PATIENTS WITH SCA AND CONTROL SUBJECTS
FVC, % pred FEV l' % pred FEV1/FVC, 8/0 FEF 2s - 75 , % pred TLC, % control RVITLC, % t Vo 2max, ml/min/kg VAT, % Vo 2 max VAT, ml/min/kg
Patients with SCA
Control Subjects
93 ± 11* 89± 13*
105 ± 13 104 ± 12
86
8186 23 30.1 69 20.5
± ± ± ± ± ±
87 95 ± 23
19§
11 23 42.4 62 26.9
4 6.6 9 5.5§
± 5 ± 7.9* ± 11 ± 7.7
Definition of abbreviations: SCA = sickle cell anemia; FEF2!lr-75 = forced expiratory flow during the middle half of the FVC; RV residual volume; Vo2 max = maximal whole-body uptake of O2 ; VAT = ventilatory anaerobic threshold. * Based on normal values of Hsu et af. (9}. t TLC expressed as percent predicted of regression equation based on height'in same-sexed control subjects (no black population normal values exist for TLC). :t p < 0.0003. §p< 0.015.
=
no difference between them in terms of maximal VTas a percent of FVC (40.9 ±. 8.1 versus 39.0 ± 7.5). Similarly, VEmax as a proportion of maximal voluntary ventilation (MVV) did not differ between the groups (0.673 ± 0.120 versus 0.592 ± 0.169, control subjects and patients, respectively; p = 0.06). The Vo2max was much lower in the patients than in the control subjects (table 2). There was a weak correlation between absolute V 90070 in all but one patient, whereas arterialized capillary P0 2 averaged 77.5 ± 8.1 mm Hg (range, 60 to 94 mm Hg), with three of 34 having a P0 2 < 70 mm Hg. The parameters used to characterize the ventilatory response to exercise for patients and control subjects are listed in table 3. The ~ VEl A Vco 2over the subVAT range of incremental exercise was significantly higher in patients with SCA than in control subjects, as was the steady-state VE/Vc02. The PETC02 during steady-state exercise did not differ between patients and control subjects (38.4 ± 3.0 versus 39.3 ± 3.2, p = 0.38), indicating similar alveolar ventilation. The Pac02 exceeded simultaneous PETC02 in only four patients by 2.25 ± 0.44 mm Hg (mean ± SD). The VD/VT ratio was
TABLE 3
PARAMETERS OF THE EXERcrSE VENTILATORY RESPONSE IN PATIENTS WITH SCA AND IN CONTROL SUBJECTS Patients with SCA HbSS AVElAVC02 ssVElVC0 2 VONT" % VT, % of FVC PETc0 2 , mm Hg Paco2 , mm Hg
30.0 ± 7.7 34.1 ± 3.6* 21.0 ± 3.0 30.5 ± 4.9 37.1 ± 3.0 34.5 ± 2.6
HbSC 28.3 32.3 19.9 28.4 39.4 37.6
± 5.2 ± 3.3 ± 3.6 ± 3.9 ± 2.5 ± 2.1§
All
Control Subjects
29.4 ± 6.5* 33.2 ± 3.5t 20.4 ± 3.3*
29.4
%
4.5
38.4 ± 3.0
23.5 30.8 17.8 31.2 39.3
± 5.6 ± 3.5 ± 2.3 ± 5.5 ± 3.2
Definition of abbreviations: SCA = sickle cell anemia; AVElAVC02 = slope of the change in expired minute ventilation versus the change in CO2 production; ss = steady state; VONT = ratio of dead space to tidal volume. * p z 0.025, compared with control subjects. f p < 0.05, compared with control subjects. p < 0.01, compared with control subjects. § p = 0.002, HbSS compared with HbSC.
*
slightly but significantly higher in patients than in control subjects, although there was no significant difference in their VT expressed as a percentage ofFVC. The findings were unchanged when only those patients with SCA who had normal spirometry were considered. When the patients with SCA were divided into those with HbSS (generally lower Hb levels) and those with HbSC, there was no difference between them comparing aVEl ~ Vco 2, which was significantly higher than in control subjects. The same was found for VD/VT, whether Pac02 or P~TC92 was used. However, steady-state VE/Vco2 in subjects with HbSC was intermediate between that of control subjects and patients with HbSS. The Pac02 was significantly lower in the latter group, indicating greater alveolar ventilation, with no difference in arterial pH between patients with HbSS and those 'Yith.HbSC. Multiple linear regression of VE/Vc0 2 on VD/VT, Hb (giL), and age revealed a positive correlation with VD/VT (figure 1) and a negative correlation with Hb concentration and age (r = 0.823). The patients with more severe clinical sickle cell disease were compared with those with a milder course. Seven subjects were identified in the former group, six with HbSS and one with HbSC, with Hb ranging from 72 to 113 giL. There was no difference in VD/VT (based on ~aco2) between the groups (more severe, 0.190 ± 0.045; less severe, 0.163 ± 0.047; p = 0.23). Neither Vo2max (28.1 ± 4.0 versus 31.1 ± 6:5 ml/min/kg, p = 0.16) nor VAT as a 07oVo2max (72070 versus 69070, p = 0.38) differed between those with severe or mild illness. In only one of the patients with SCA who was considered to have more severe clinical disease did Paeo, exceed PETC02. Discussion This study has shown that children with SCA have an exaggerated ventilatory response to exercise because of a combination of factors, most important of which are increased physiologic dead space and low hemoglobin level. These findings are in agreement with those of Miller and coworkers (1) who demonstrated a greater ventilatory response in adults with SCA and hypocapnia in the more severely anemic patients. The calculation of physiologic VD (Vnphys) requires subtraction of the apparatus dead space (vnapp), but no such correction was made for VT. In larger subjects with large VT, small errors in esti-
229
VENTILATION AND GAS EXCHANGE DURING EXERCISE IN SICKLE CELL ANEMIA
mation of Vnapp will have little impact on measurement of Vnphys, but the same is not true for smaller subjects with smaller VT. Although this would lead to negligible errors in the calculation of the Vn/VT ratio in larger subjects, relatively larger errors may occur in smaller subjects (9). However, because the patients with SCA and the control subjects were closely matched for height and weight, and because there was no age effect on the Vn/VT ratio, this should not have affected the findings in this study. The slightly lower spirometric values seen in the patients with SCA are in keeping with studies in adults with this disease (1). It is conceivable that this could have contributed to the higher Vn/VT ratio seen in this study, if one supposes that VT recruitment was limited by smaller lung volumes. This is unlikely to be the case because the strategy for tapping into the ventilatory reserve in response to exercise was identical between groups, in agreement with findings in adults with SCA (1). Hypoxemia was demonstrated in some patients during exercise. Po, obtained from a warmed capillary bed in exercise yields values of P0 2 similar to arterial P0 2 (10). The mean arterialized capillary P0 2 in the patients was 77.5 mm Hg, very close to the Pa02 obtained at rest by Wall and coworkers (3) in children with SCA. Arterial saturation in SCA may be lower than in normal subjects because of the right-shifted hemoglobin-oxygen dissociation curve, even in the face of normal or minimally reduced Pa02(11). This suggests that saturations between 90 and 950/0, seen in one-quarter ofthe subjects, represents (at most) mild hypoxemia. There was no correlation between Pa02 and the ventilatory equivalent for CO 2. The effect of anemia (viz., hypoxia) on the ventilatory response to exercise is not well understood. In two studies, irondeficient anemic subjects, with Hb levels as low as 67 ± 11 g/L, had similar ventilatory equivalents to those of nonanemic control subjects (12, 13). In contrast, Andersen and Barkve (14) demonstrated a higher ventilatory equivalent for O 2in iron-deficient anemic subjects (Hb as low as 77 g/L), especially at higher work loads, which fell with treatment of the anemia. In none of these studies were blood gases measured, nor was reference made to alveolar ventilation. The patients with HbSS in the present study had lower Pac02 in the absence of acidosis. A lower CO 2 set-point implies greater ventilation for a given metabolic load (15),
35
Fig. 1. Plot of steady-slate ventilatory equivalent for CO 2 (VE1VcxJ,) versus VoM for all patients with SCA. Calculation of VO based on measured Paco2 • Patients with HbSS represented by solid circles; those with HbSC by open circles.
N
o u
.>
3 0
2J ,.----,..------,,---- ----,-----, 0. 05
but the reason for a lower set-point in Unlike the normal red cell, which deforms anemic subjects (if it exists) is not known. easily and passes quickly through capilThere is no evidence that the response laries, the sickled cell is much less deforto inhaled CO 2 is different in patients mable and may pass much more slowly with SCA from that in normal subjects through the pulmonary capillaries (23). (16). However, because Hb plays a ma- Because this functional delay is influjor role in buffering the hydrogen ion enced by Po 2, it would be less evident at from carbonic acid, a normal pH in ane- . rest when venous saturations are relativemia would be expected to require a lower ly high but would increase with exercise Paco2 (17, 18); There was no correlation because of the inflow of markedly between Hb level and Pan, in patients desaturated blood to the lungs from the with SCA. The possibility that an aug- exercising limbs (20). It would be reversmented exercise ventilatory response ible upon recovery from exercise. Alresults from elevated cardiac output and though the number of "more severely afbaroreceptor activation (19) in a chroni- fected" patients with SCA was small, cally vasodilated, high output state such there was no difference between their as SCA (20) must be considered. The aorVn/VT ratio and that of their less severetic body chemoreceptors, whose role in ly affected peers. This is in line with aurespiratory regulation is considered topsy findings in adults with SCA, which negligible (6), may be involved in this re- reveal a remarkable paucity of obliteragard because they have been shown to tive pulmonary vascular disease. This would suggest that if the "acute chest synrespond to O 2 content (21). The normal ventilatory response to ex- drome" is an intrapulmonary vaso-occluercise in children has been described by sive phenomenon, it does not usually lead Cooper and coworkers (22). The single to progressive obstruction of the pulmomost important factor determining ven- nary circulation. tilation was the metabolic load expressed . It is remarkable that despite all the as VC02' The other determinant was age, pathophysiologic alterations in cardiopuli.e., younger children required greater monary function described herein and overall ventilation than did older children, elsewhere (4), exercise tests were completpresumably because they regulate their ed without any deleterious effects on the Paoo, around a somewhat lower set-point patients with SCA. This study would sug(22). In addition, in SCA the greater dead gest that short bursts of vigorous exerspace may mean a greater overall venti- cise are safe provided the child stops when lation. Miller and coworkers (1) found his or her own point of exhaustion is that Vn in adults with SCA was no differ- reached. Caution must be exercised in hot ent than in control subjects at rest, but weather, when access to fluids is limited, they reported that during exercise the and for prolonged submaximal exercise Vn/VT ratio did not fall from its resting until further data are available. level in some patients. This is compatible In conclusion, children with sickle cell with intrapulmonary sickling being a dy- anemia have an exaggerated ventilatory namic process and not caused by irrevers- response to exercise because of increased ible obliteration of the pulmonary vessels. physiologic dead space and because of
230
PIANOSI, D'SOUZA, ESSELTINE, CHARGE, AND COATES
their low Hb concentration. As a group they maintain normal alveolar ventilation; however, those with more severe anemia hyperventilate and develop hypocapnia despite greater dead space. The most likely explanation for the greater dead space in SeA is intrapulmonary sickling where red cells impair perfusion of capillaries in normally ventilated alveoli. The effect of anemia per se on the ventilatory response to exerciserequires further study. References 1. Miller GJ, Serjeant GR, Sivapragasam A, Petch MC. Cardiopulmonary responsesand gas exchange during exercisein adults with sicklecelldisease(sickle cell anaemia). Clin Sci 1973; 44:113-28. 2. Bromberg PA, Jensen WN. Arterial oxygenunsaturation in sickle cell disease. Am Rev Respir Dis 1967; 96:400-7. 3. Wall MA, Platt OS, Strieder DJ. Lung function in children with sickle cell anemia. Am Rev Respir Dis 1979; 120:210-4. 4. Alpert BS, Dover EV, Strong WB, Covitz W. Longitudinal exercise hemodynamics in children with sickle cell anemia. Am J Dis Child 1984; 138:1021-4. 5. Jones NL. Clinicalexercise testing. 3rd ed. Montreal: WB Saunders, 1988; 135-44, 289-300. 6. Wasserman K, Hansen JE, Sue DY,Whipp BJ.
Principles of exercise testing and interpretation. Philadelphia: Lea and Febiger, 1987; 27-46. 7. Grant GP, Graziano JH, Seaman C, Mansell AL. Cardiorespiratory response to exercise in patients with thalassemia major. Am Rev Respir Dis 1987; 136:92-7. 8. Minitab data analysis software, 1984.State College, PA: Minitab, Inc., 1984. 9. Hsu KHK, Jenkins DE, Hsi BP, et ale Ventilatory functions of normal children and young adults - Mexican-American, white, and black. I. Spirometry. J Pediatr 1979; 95:14-23. 10. Godfrey S. Exercise testing in children. Philadelphia: WB Saunders, 1974; 25-6. 11. Becklake MR, Griffiths SB, McGregor M, Gordman HI, Schreve JP. Oxygen dissociation curves in sickle cell anemia and in subjects with the sickle cell trait. J Clin Invest 1955; 34:751-5. 12. Davies CTM, Chukweumeka AC, Van Haaren JPM. Iron deficiencyanaemia: its effect on maximum aerobic power and responses to exercise in African males aged 17-40 years. Clin Sci 1973; 44:555-62. 13. Cotes JE, Dabbs JM, Elwood PC, Hall AM, McDonald A, Saunders MJ. Iron-deficiency anaemia: its effect on transfer factor for the lung (diffusing capacity) and ventilation and cardiac frequency during sub-maximal exercise. Clin Sci 1972; 42:325-35. 14. Andersen HT, Barkve H. Iron deficiency and muscular work performance. Scand J Clin Lab Invest [Suppl] 1970; 114:22-6. 15. Oren A, Wasserman K, Davis JA, Whipp BJ. Effect of CO 2 set-point on ventilatory response to exercise. J Appl Physiol 1981; 51:185-9.
16. Elegbeleye DO, Akinsete FI, Afonja AO, FemiPearse D. Ventilatory response to carbon dioxide in patients with homozygous sickle cell disease. Br J Anaesthesiol 1976; 48:249-52. 17. Ueda Y, Bookchin RM. Effects of carbon dioxide levelpH variations in vitro on blood respiratory functions, red blood cell volume, transmembrane pH gradients and sickling in sickle cell anemia. J Lab Clin Med 1984; 104:146-59. 18. Ohira Y, Simpson DR, Edgerton VR, Gardner GW, Senewiratne B. Characteristics of blood gas in response to iron treatment and exercise in iron-deficient and anemic subjects. J Nutr Sci Vitaminol (Tokyo) 1983; 29:129-39. 19. Jones PW, Huszczuk A, Wasserman K. Cardiac output as a controller of ventilation through changes in right ventricular load. J Appl Physiol 1982; 53:218-24. 20. Lonsdorfer J, Bogui P, Otayeck A, Bursaux E, Poyart C, Cabannes R. Cardiorespiratory adjustments in chronic sickle cell anemia. Bull Eur Physiopathol Respir 1983; 19:339-44. 21. MitchellRA, Hatcher JD, GillisD. Discussion. Chemoreflexes and the circulation. In: Purves MJ, ed. The peripheral arterial chemoreceptors. London: Cambridge University Press, 1975; 445-6. 22. Cooper DM, Kaplan MR, Baumgarten L, Weiller-RavellD, Whipp BJ, Wasserman K. Coupling of ventilation and CO 2 production during exercise in children. Pediatr Res 1987; 21:568-72. 23. Vayo MM, Lipowsky HH, Karp N, Schmalzer E, Chein S. A model of microvascular oxygen transport in sickle cell disease. MicrovascRes 1985; 30:195-206.
Introduction
Sickle cell anemia (SCA) is characterized by the propensity for HbS to polymerize in a hypoxic milieu, resulting in changes in the shape and deformability of the red blood cell. In physiologicterms, occlusion of the pulmonary capillaries would result in a region of ventilationl perfusion fY/Q) imbalance, not unlike that seen in pulmonary embolism, leading to increased dead space. This could be a permanent vascular occlusion leading to infarction or a functional occlusion where red cell movement is slowed but not completely arrested. Older patients, especially those who have experienced multiple episodes of "acute chest syndrome," are presumably more at risk for permanent impairment than are younger patients. Elevation of dead space has been found in adults with SCA during exercise (1) but not at rest. In order to achieve the level of alveolar ventilation required to maintain normocapnia in the face of increased dead space, overall ventilation is increased (1). However, there are other recognized physiologic derangements in sickle cell disease that might account for the hyperventilation seen during exercise. Miller and coworkers (1) concluded that in addition to greater dead space ventilation, an increased need for anaerobic metabolism in the more anemic patients and the hypoxemia seen in this disorder (2) contributed to the exaggerated ventilatory response to exercise. In children with SCA, Wall and coworkers (3) found mild hypoxemia but spirometric indices similar to those in control subjects. Alpert and associates (4) have demonstrated low exercisetolerance, and studied cardiovascular responses to exercise in these children, but there have been no studies on the ventilatory response to exercisein this group. The purpose of this investigation was to measure the ventilatory response to exercise and to assess the roles of increased physiologic dead space, anaerobic metabolism, and hypoxemia in determining this response in a group of children with SCA that included those who had suffered episodes of "acute chest syn226
SUMMARY Adults with sickle cell anemia (SCA) have restrictive lung Impairment, Increased alveolar dead space, and hypoxemia. These factors, together with Increased anaerobic metabolism, are thought to cause exercise hyperventilation. To assess the role of each of these In children, 34 patients with SCA and 16 control subjects performed pulmonary function and exercise tests. 1Wentyeight patients with SCA had spirometric values and lung volumes, and all but two patients with SCA had arterial saturation> 91% during exercise. Despite a low Vo 2 max (30.07 ± 6.55 ml/mln/kg), the ventilatory anaerobic threshold (VAT) In the patients occurred at a similar % Vo 2max as In the control subjects (69 ± 9% versus 63 ± 12%). The slope of the A VEl AVC0 2 relationship for sub-VAT work was steeper In the patients (29.4 ± 6.5 versus 24.7 ± 5.2, P = 0.01),and the ventilatory equivalent for CO2 (VElVC0 2) In steady-state exercise was greater In the patients than In the control sub0.03). End-tidal PC0 2 did not differ (38.3 ± 3.0 versus Jects (33.2 ± 3.5 versus 30.8 ± 3.5, P 39.2 ± 3.1), Indicating equivalent alveolar ventilation. The patients had a higher dead space:tldal volume ratio (VDNT) than did the control subjects (0.204 ± 0.033 versus 0.173 ± 0.024, P = 0.0005). The PaC0 2 was significantly lower In those with lower Hb, but there was no difference In pH. In conclusion, children with SCA have an Increased exercise ventilatory response caused In part by Increased physiologic dead space, and In part by their low Hb. The greater dead space may be the result of sickle cells Impairing capillary perfusion to ventilated alveoli.
=
AM REV RESPIR DIS 1991; 143:228-230
drome" and those who had experienced a more benign course. Methods Study Population Subjects were recruited from the Sickle Cell Anemia Clinic at the Montreal Children's Hospital. Inclusion criteria were: age, 6 to 19 yr; height> 120 em (lower height limit of the exercise cycle ergometer), HbSS or HbSC on hemoglobin electrophoresis, absence of congenital heart disease or hypertension, and not participating in a regular transfusion program. Patients were studied a minimum of 4 wk after any illness, and 3 months after any hospitalization. All patients were studied in their normal stable condition as judged by hemoglobin level and clinical assessment. Control subjects were recruited from the families of hospital personnel or from siblings of the patients with SeA (HbAA), and they were free of cardiopulmonary disease, The hemoglobin level and type were known in all siblings, and screening was offered to nonrelated control subjects. The study received approval from the Institutional Review Board, and all subjects gave informed, written consent. The charts were reviewed by one author (SJAD'S), and those responsible for the exercise testing had no previous knowledge of the results. Patients who exhibited a more severe clinical course were compared with thei~ counterparts with milder disease. A more severeclinicalcourse wasdefined as having been
hospitalized, or having received an exchange transfusion, for treatment of one or more episodes of "acute chest syndrome," or having been admitted to the ICU for any reason related to management of a vaso-occlusive crisis. It was hypothesized that such subjects would be at highest risk of developing irreversible changes in their pulmonary vasculature because of the sickling process, which would be made manifest by an elevated VD/VT ratio.
Protocol All testing was done the same day (morning or afternoon) after a light meal. Subjects had electrocardiograms to look for possible arrhythmias and then performed pulmonary function testing consisting 0 f spirometry and
(Received in originalform September 22, 1989and in revised form May 18, 1990) 1 From the Respiratory Medicine and Hematology Services, McGill University-Montreal Children's Hospital Research Institute, Montreal, Quebec, Canada. 2 Publication no. 90-023 of the McGill University-Montreal Children's Hospital Research Institute. 3 Correspondence and requests for reprints should be addressed to Dr. Allan L. Coates, The Montreal Children's Hospital, 2300 Tupper Street, Montreal, Quebec H3H IP3, Canada. 4 Fellow of the Canadian Cystic Fibrosis Foundation. S Funded by Le Fonds de La Recherche en Sante du Quebec.
VENTILATION AND GAS EXCHANGE DURING EXERCISE IN SICKLE CELL ANEMIA
plethysmographic lung volume measurement. They then performed a progressive exercise test (Jones Stage I) (5) on an electronically braked cycle ergometer (Quinton Uniwork Model 884 for those between 120and 150cm; Lode Corival 300 for those> 150 em) to exhaustion. The final load was taken as their maximal work load (Wmax). The increment in work load each minute varied according to the size of the subject, being 8 W for the smallest children and as much as 20 W for adolescents, such that the test would be expected to last from 5 to 10 min. After a l-h rest period, during which subjects usually had a snack, they pedaled at 500/0 of their Wmax at a frequency of 50 to 70 rpm for 5 to 7 min (Jones Stage III). The work load of 500/0 Wmax was chosen because past experience had shown that not all untrained or sick subjects were capable of maintaining a work load of two-thirds Wmax, the usual standard load, for the time necessary to achieve good steadystate values. Steady-state conditions were considered present when fluctuations of not more than ± 0.10/0 in mixed expired gas concentrations and ± 5 bpm in heart rate lasting at least 30 s were noted after the third minute of exercise (5). At this time, arterialized capillary blood specimens consisting of two or three 125-J.11 tubes were obtained. In two subjects the samples wererejected because of poor blood flow from the finger-prick. No blood sampling was done on control subjects during exercise. During exercise, subjects wore noseclips and breathed through a two-way, nonrebreathing valve(Hans Rudolph, Kansas City, MO or Koegel, Houston, TX). Apparatus dead space varied with subject size but was generally set up as follows: 55 ml for subjects < 140 em, 65 ml for subjects 140 to 160 em, and 75 ml for subjects> 160em, Although the resistance of the valves also varied, being greater with the smaller apparatus, size-related variability in flow rates ensured that resistance remained negligible.
Measurements Heart rate (HR) and saturation were monitored using a pulse oximeter (Model N-2oo; Nellcor, Hayward, CA). Inspired ventilation was measured with a dry gas meter (Parkinson-Cowan, Manchester, UK). Expired gas was sampled from a mixing chamber with a fan, the volume of which was adjusted to approximately 750/0 of the subject's vital capacity. The expirate passed through 10to 20 mesh CaS04 (Drierite) to remove moisture, and the concentrations of O 2 and CO 2 (FE02, FEc0 2) were measured by rapidly responding gas analyzers (Model SA-3; Applied Electrochemistry, Pasadena, CA, and Model LB-2; Beckman Instruments, Fullerton, CA). Both were calibrated with room air and a test gas of known concentration (gravimetric method) before each stage of exercise testing. End-tidal CO 2 concentration was also sampled via a heated line at the mouthpiece over 3 to 6 breaths during the latter half of each minute, Output signals for HR, VI, FE02, FEC02' and FETC02 wererecorded on an eight-
channel chart recorder (Model 7758A; Hewlett-Packard, Waltham, MA), whereas the numerical value for saturation was recorded manually three to five times per minute. From the inspired minute ventilation (VI), FE02, and FEc0 2, measured over the final 15 to 20 s of each minute, expired minute ventilation (VE), oxygen consumption (V02), and carbon dioxide production (VC02) were calculated by standard methods (5) and considered as the values representative at that work load. Valuesfor VE werecorrected to BTPS conditions, whereas V02and VC02 were expressed as STPn values. The V02for the last work load was taken as the subject's V02max. The ventilatory anaerobic threshold (VAT) was determined by the V-slope method (6), and it was described both in terms of absolute V02 (ml/min/kg) and as % V02max. Blood specimens were stored on ice until analyzed for pH, P0 2, and Pc0 2 (ABL3 blood gas analyzer; Radiometer, Copenhagen, Denmark) within 20 min of sampling.
Analysis A test was considered acceptable if there was a progressive increase in VE, V02, and VC02. Transient increases in ventilation were felt to represent hyperventilation, and tests with those types of changes were rejected. The ventilatory response to exercise was assessed in two ways. First, the slope of the change in VE versus the change in VC02 (AVElAVC02) was calculated over the sub-VAT range of incremental exercise, and then ventilatory equivalent for CO 2during steady-state exercise(VEIVc02) was calculated. Dead space (Vo) was calculated using the Bohr equation using Pe02values obtained in steady-state exercise, with the appropriate correction made for apparatus dead space. Because Vn increases with increasing tidal volume (VT), and to facilitate comparison between different-sized subjects, the ratio of dead space to tidal volume (Vn/VT) was examined. As blood sampling was not done on control subjects, the PETC02 was used as an estimate of Pac02 in the calculation of Vn/VT in this group. Thus, when the groups werecompared, the calculated value for both control subjects and patients with SCA was based on PETC02. This was not unreasonable since the difference between simultaneously obtained PETC02 and Pac02 in the patients with SCA was 2.25 ± 2.27 mm Hg (range, +5.6to -2.7). Grant and coworkers (7) cite similar differences that wereinsignificant, and the use of PETC02' even in the presence of lung disease, appears to be more reliable than previously thought (7). The relation between the VE/VC02 and VD/VT in patients with SCA during steady-state exercise was explored by linear regression techniques, and in this case the values for Vn/VT were those based on Pac02. Comparisons between patients and control subjects were tested for statistical significance using the two-tailed unpaired t test, with a < 0.05. Statistical analysis was done using a Minitab PC program (8).
227
Results
A total of 50 successful studies were accomplished. There were three patients with SCA and three control subjects for whom either hyperventilation, lack of effort, or technical problems were grounds for excluding the tests from further analysis. Of the 34 patients with SCA, 18 had HbSS, with levels ranging from 65 to 106 giL and 16had HbSC, with levelsranging from 109 to 133 giL. This represented 80070 of patients in the clinic who met the inclusion criteria. Of those studied, 25070 were considered to have severe clinical disease; the same proportion among those not studied was 20070. Sixteen control subjects were studied, of whom nine were documented to have HbAA; the remaining seven declined a screening test. Demographic data are displayed in table 1, which shows that patients with SCA and control subjects were very closely matched for sex, age, height, and weight. The main difference between the groups was hemoglobin levels. Pulmonary function data are listed in table 2. An in-depth report of these will appear in a separate article, but it is apparent from the table that patients with SCA had lower spirometric indices than did control subjects. Twenty-eight of the patients had spirometric indices within normal limits (9), whereas all control subjects were within the accepted normal range. All patients with SCA completed the protocol without incident. All but one gave what appeared to be maximal performance as judged by effort and peak HR ~ 174. One subject had a peak HR of 156, but his effort was judged to be maximal nonetheless, whereas another achieved a peak HR of 172, but her effort appeared submaximal, and therefore her Vo2 max data were excluded from analysis.Ventilation at maximal exercisewas greater in control subjects than in patients with SCA (1.396 ± 0.285 versus 1.204 ± 0.287Llminlkg, p = 0.03), but there was
TABLE 1 PHYSICAL CHARACTERISTICS OF PATIENTS WITH SCA AND CONTROL SUBJECTS· Patients M/F Age, yr Height, cm Weight, kg Hb, gIL
17/17
13.0 153 44 101
± 3.8 ± 15 ± 15 ± 18
Control Subjects
8/8 12.3 ± 3.5 152 ± 15 46 ± 16
Definitionof abbreviations: SeA = sickle cell anemia; Hb = hemoglobin. * Values are mean ± SO.
228
PIANOSI, D'SOUZA, ESSELTINE, CHARGE, AND COA'!'ES
TABLE 2 PULMONARY FUNCTION* AND AEROBIC EXERCISE PARAMETERS OF PATIENTS WITH SCA AND CONTROL SUBJECTS
FVC, % pred FEV l' % pred FEV1/FVC, 8/0 FEF 2s - 75 , % pred TLC, % control RVITLC, % t Vo 2max, ml/min/kg VAT, % Vo 2 max VAT, ml/min/kg
Patients with SCA
Control Subjects
93 ± 11* 89± 13*
105 ± 13 104 ± 12
86
8186 23 30.1 69 20.5
± ± ± ± ± ±
87 95 ± 23
19§
11 23 42.4 62 26.9
4 6.6 9 5.5§
± 5 ± 7.9* ± 11 ± 7.7
Definition of abbreviations: SCA = sickle cell anemia; FEF2!lr-75 = forced expiratory flow during the middle half of the FVC; RV residual volume; Vo2 max = maximal whole-body uptake of O2 ; VAT = ventilatory anaerobic threshold. * Based on normal values of Hsu et af. (9}. t TLC expressed as percent predicted of regression equation based on height'in same-sexed control subjects (no black population normal values exist for TLC). :t p < 0.0003. §p< 0.015.
=
no difference between them in terms of maximal VTas a percent of FVC (40.9 ±. 8.1 versus 39.0 ± 7.5). Similarly, VEmax as a proportion of maximal voluntary ventilation (MVV) did not differ between the groups (0.673 ± 0.120 versus 0.592 ± 0.169, control subjects and patients, respectively; p = 0.06). The Vo2max was much lower in the patients than in the control subjects (table 2). There was a weak correlation between absolute V 90070 in all but one patient, whereas arterialized capillary P0 2 averaged 77.5 ± 8.1 mm Hg (range, 60 to 94 mm Hg), with three of 34 having a P0 2 < 70 mm Hg. The parameters used to characterize the ventilatory response to exercise for patients and control subjects are listed in table 3. The ~ VEl A Vco 2over the subVAT range of incremental exercise was significantly higher in patients with SCA than in control subjects, as was the steady-state VE/Vc02. The PETC02 during steady-state exercise did not differ between patients and control subjects (38.4 ± 3.0 versus 39.3 ± 3.2, p = 0.38), indicating similar alveolar ventilation. The Pac02 exceeded simultaneous PETC02 in only four patients by 2.25 ± 0.44 mm Hg (mean ± SD). The VD/VT ratio was
TABLE 3
PARAMETERS OF THE EXERcrSE VENTILATORY RESPONSE IN PATIENTS WITH SCA AND IN CONTROL SUBJECTS Patients with SCA HbSS AVElAVC02 ssVElVC0 2 VONT" % VT, % of FVC PETc0 2 , mm Hg Paco2 , mm Hg
30.0 ± 7.7 34.1 ± 3.6* 21.0 ± 3.0 30.5 ± 4.9 37.1 ± 3.0 34.5 ± 2.6
HbSC 28.3 32.3 19.9 28.4 39.4 37.6
± 5.2 ± 3.3 ± 3.6 ± 3.9 ± 2.5 ± 2.1§
All
Control Subjects
29.4 ± 6.5* 33.2 ± 3.5t 20.4 ± 3.3*
29.4
%
4.5
38.4 ± 3.0
23.5 30.8 17.8 31.2 39.3
± 5.6 ± 3.5 ± 2.3 ± 5.5 ± 3.2
Definition of abbreviations: SCA = sickle cell anemia; AVElAVC02 = slope of the change in expired minute ventilation versus the change in CO2 production; ss = steady state; VONT = ratio of dead space to tidal volume. * p z 0.025, compared with control subjects. f p < 0.05, compared with control subjects. p < 0.01, compared with control subjects. § p = 0.002, HbSS compared with HbSC.
*
slightly but significantly higher in patients than in control subjects, although there was no significant difference in their VT expressed as a percentage ofFVC. The findings were unchanged when only those patients with SCA who had normal spirometry were considered. When the patients with SCA were divided into those with HbSS (generally lower Hb levels) and those with HbSC, there was no difference between them comparing aVEl ~ Vco 2, which was significantly higher than in control subjects. The same was found for VD/VT, whether Pac02 or P~TC92 was used. However, steady-state VE/Vco2 in subjects with HbSC was intermediate between that of control subjects and patients with HbSS. The Pac02 was significantly lower in the latter group, indicating greater alveolar ventilation, with no difference in arterial pH between patients with HbSS and those 'Yith.HbSC. Multiple linear regression of VE/Vc0 2 on VD/VT, Hb (giL), and age revealed a positive correlation with VD/VT (figure 1) and a negative correlation with Hb concentration and age (r = 0.823). The patients with more severe clinical sickle cell disease were compared with those with a milder course. Seven subjects were identified in the former group, six with HbSS and one with HbSC, with Hb ranging from 72 to 113 giL. There was no difference in VD/VT (based on ~aco2) between the groups (more severe, 0.190 ± 0.045; less severe, 0.163 ± 0.047; p = 0.23). Neither Vo2max (28.1 ± 4.0 versus 31.1 ± 6:5 ml/min/kg, p = 0.16) nor VAT as a 07oVo2max (72070 versus 69070, p = 0.38) differed between those with severe or mild illness. In only one of the patients with SCA who was considered to have more severe clinical disease did Paeo, exceed PETC02. Discussion This study has shown that children with SCA have an exaggerated ventilatory response to exercise because of a combination of factors, most important of which are increased physiologic dead space and low hemoglobin level. These findings are in agreement with those of Miller and coworkers (1) who demonstrated a greater ventilatory response in adults with SCA and hypocapnia in the more severely anemic patients. The calculation of physiologic VD (Vnphys) requires subtraction of the apparatus dead space (vnapp), but no such correction was made for VT. In larger subjects with large VT, small errors in esti-
229
VENTILATION AND GAS EXCHANGE DURING EXERCISE IN SICKLE CELL ANEMIA
mation of Vnapp will have little impact on measurement of Vnphys, but the same is not true for smaller subjects with smaller VT. Although this would lead to negligible errors in the calculation of the Vn/VT ratio in larger subjects, relatively larger errors may occur in smaller subjects (9). However, because the patients with SCA and the control subjects were closely matched for height and weight, and because there was no age effect on the Vn/VT ratio, this should not have affected the findings in this study. The slightly lower spirometric values seen in the patients with SCA are in keeping with studies in adults with this disease (1). It is conceivable that this could have contributed to the higher Vn/VT ratio seen in this study, if one supposes that VT recruitment was limited by smaller lung volumes. This is unlikely to be the case because the strategy for tapping into the ventilatory reserve in response to exercise was identical between groups, in agreement with findings in adults with SCA (1). Hypoxemia was demonstrated in some patients during exercise. Po, obtained from a warmed capillary bed in exercise yields values of P0 2 similar to arterial P0 2 (10). The mean arterialized capillary P0 2 in the patients was 77.5 mm Hg, very close to the Pa02 obtained at rest by Wall and coworkers (3) in children with SCA. Arterial saturation in SCA may be lower than in normal subjects because of the right-shifted hemoglobin-oxygen dissociation curve, even in the face of normal or minimally reduced Pa02(11). This suggests that saturations between 90 and 950/0, seen in one-quarter ofthe subjects, represents (at most) mild hypoxemia. There was no correlation between Pa02 and the ventilatory equivalent for CO 2. The effect of anemia (viz., hypoxia) on the ventilatory response to exercise is not well understood. In two studies, irondeficient anemic subjects, with Hb levels as low as 67 ± 11 g/L, had similar ventilatory equivalents to those of nonanemic control subjects (12, 13). In contrast, Andersen and Barkve (14) demonstrated a higher ventilatory equivalent for O 2in iron-deficient anemic subjects (Hb as low as 77 g/L), especially at higher work loads, which fell with treatment of the anemia. In none of these studies were blood gases measured, nor was reference made to alveolar ventilation. The patients with HbSS in the present study had lower Pac02 in the absence of acidosis. A lower CO 2 set-point implies greater ventilation for a given metabolic load (15),
35
Fig. 1. Plot of steady-slate ventilatory equivalent for CO 2 (VE1VcxJ,) versus VoM for all patients with SCA. Calculation of VO based on measured Paco2 • Patients with HbSS represented by solid circles; those with HbSC by open circles.
N
o u
.>
3 0
2J ,.----,..------,,---- ----,-----, 0. 05
but the reason for a lower set-point in Unlike the normal red cell, which deforms anemic subjects (if it exists) is not known. easily and passes quickly through capilThere is no evidence that the response laries, the sickled cell is much less deforto inhaled CO 2 is different in patients mable and may pass much more slowly with SCA from that in normal subjects through the pulmonary capillaries (23). (16). However, because Hb plays a ma- Because this functional delay is influjor role in buffering the hydrogen ion enced by Po 2, it would be less evident at from carbonic acid, a normal pH in ane- . rest when venous saturations are relativemia would be expected to require a lower ly high but would increase with exercise Paco2 (17, 18); There was no correlation because of the inflow of markedly between Hb level and Pan, in patients desaturated blood to the lungs from the with SCA. The possibility that an aug- exercising limbs (20). It would be reversmented exercise ventilatory response ible upon recovery from exercise. Alresults from elevated cardiac output and though the number of "more severely afbaroreceptor activation (19) in a chroni- fected" patients with SCA was small, cally vasodilated, high output state such there was no difference between their as SCA (20) must be considered. The aorVn/VT ratio and that of their less severetic body chemoreceptors, whose role in ly affected peers. This is in line with aurespiratory regulation is considered topsy findings in adults with SCA, which negligible (6), may be involved in this re- reveal a remarkable paucity of obliteragard because they have been shown to tive pulmonary vascular disease. This would suggest that if the "acute chest synrespond to O 2 content (21). The normal ventilatory response to ex- drome" is an intrapulmonary vaso-occluercise in children has been described by sive phenomenon, it does not usually lead Cooper and coworkers (22). The single to progressive obstruction of the pulmomost important factor determining ven- nary circulation. tilation was the metabolic load expressed . It is remarkable that despite all the as VC02' The other determinant was age, pathophysiologic alterations in cardiopuli.e., younger children required greater monary function described herein and overall ventilation than did older children, elsewhere (4), exercise tests were completpresumably because they regulate their ed without any deleterious effects on the Paoo, around a somewhat lower set-point patients with SCA. This study would sug(22). In addition, in SCA the greater dead gest that short bursts of vigorous exerspace may mean a greater overall venti- cise are safe provided the child stops when lation. Miller and coworkers (1) found his or her own point of exhaustion is that Vn in adults with SCA was no differ- reached. Caution must be exercised in hot ent than in control subjects at rest, but weather, when access to fluids is limited, they reported that during exercise the and for prolonged submaximal exercise Vn/VT ratio did not fall from its resting until further data are available. level in some patients. This is compatible In conclusion, children with sickle cell with intrapulmonary sickling being a dy- anemia have an exaggerated ventilatory namic process and not caused by irrevers- response to exercise because of increased ible obliteration of the pulmonary vessels. physiologic dead space and because of
230
PIANOSI, D'SOUZA, ESSELTINE, CHARGE, AND COATES
their low Hb concentration. As a group they maintain normal alveolar ventilation; however, those with more severe anemia hyperventilate and develop hypocapnia despite greater dead space. The most likely explanation for the greater dead space in SeA is intrapulmonary sickling where red cells impair perfusion of capillaries in normally ventilated alveoli. The effect of anemia per se on the ventilatory response to exerciserequires further study. References 1. Miller GJ, Serjeant GR, Sivapragasam A, Petch MC. Cardiopulmonary responsesand gas exchange during exercisein adults with sicklecelldisease(sickle cell anaemia). Clin Sci 1973; 44:113-28. 2. Bromberg PA, Jensen WN. Arterial oxygenunsaturation in sickle cell disease. Am Rev Respir Dis 1967; 96:400-7. 3. Wall MA, Platt OS, Strieder DJ. Lung function in children with sickle cell anemia. Am Rev Respir Dis 1979; 120:210-4. 4. Alpert BS, Dover EV, Strong WB, Covitz W. Longitudinal exercise hemodynamics in children with sickle cell anemia. Am J Dis Child 1984; 138:1021-4. 5. Jones NL. Clinicalexercise testing. 3rd ed. Montreal: WB Saunders, 1988; 135-44, 289-300. 6. Wasserman K, Hansen JE, Sue DY,Whipp BJ.
Principles of exercise testing and interpretation. Philadelphia: Lea and Febiger, 1987; 27-46. 7. Grant GP, Graziano JH, Seaman C, Mansell AL. Cardiorespiratory response to exercise in patients with thalassemia major. Am Rev Respir Dis 1987; 136:92-7. 8. Minitab data analysis software, 1984.State College, PA: Minitab, Inc., 1984. 9. Hsu KHK, Jenkins DE, Hsi BP, et ale Ventilatory functions of normal children and young adults - Mexican-American, white, and black. I. Spirometry. J Pediatr 1979; 95:14-23. 10. Godfrey S. Exercise testing in children. Philadelphia: WB Saunders, 1974; 25-6. 11. Becklake MR, Griffiths SB, McGregor M, Gordman HI, Schreve JP. Oxygen dissociation curves in sickle cell anemia and in subjects with the sickle cell trait. J Clin Invest 1955; 34:751-5. 12. Davies CTM, Chukweumeka AC, Van Haaren JPM. Iron deficiencyanaemia: its effect on maximum aerobic power and responses to exercise in African males aged 17-40 years. Clin Sci 1973; 44:555-62. 13. Cotes JE, Dabbs JM, Elwood PC, Hall AM, McDonald A, Saunders MJ. Iron-deficiency anaemia: its effect on transfer factor for the lung (diffusing capacity) and ventilation and cardiac frequency during sub-maximal exercise. Clin Sci 1972; 42:325-35. 14. Andersen HT, Barkve H. Iron deficiency and muscular work performance. Scand J Clin Lab Invest [Suppl] 1970; 114:22-6. 15. Oren A, Wasserman K, Davis JA, Whipp BJ. Effect of CO 2 set-point on ventilatory response to exercise. J Appl Physiol 1981; 51:185-9.
16. Elegbeleye DO, Akinsete FI, Afonja AO, FemiPearse D. Ventilatory response to carbon dioxide in patients with homozygous sickle cell disease. Br J Anaesthesiol 1976; 48:249-52. 17. Ueda Y, Bookchin RM. Effects of carbon dioxide levelpH variations in vitro on blood respiratory functions, red blood cell volume, transmembrane pH gradients and sickling in sickle cell anemia. J Lab Clin Med 1984; 104:146-59. 18. Ohira Y, Simpson DR, Edgerton VR, Gardner GW, Senewiratne B. Characteristics of blood gas in response to iron treatment and exercise in iron-deficient and anemic subjects. J Nutr Sci Vitaminol (Tokyo) 1983; 29:129-39. 19. Jones PW, Huszczuk A, Wasserman K. Cardiac output as a controller of ventilation through changes in right ventricular load. J Appl Physiol 1982; 53:218-24. 20. Lonsdorfer J, Bogui P, Otayeck A, Bursaux E, Poyart C, Cabannes R. Cardiorespiratory adjustments in chronic sickle cell anemia. Bull Eur Physiopathol Respir 1983; 19:339-44. 21. MitchellRA, Hatcher JD, GillisD. Discussion. Chemoreflexes and the circulation. In: Purves MJ, ed. The peripheral arterial chemoreceptors. London: Cambridge University Press, 1975; 445-6. 22. Cooper DM, Kaplan MR, Baumgarten L, Weiller-RavellD, Whipp BJ, Wasserman K. Coupling of ventilation and CO 2 production during exercise in children. Pediatr Res 1987; 21:568-72. 23. Vayo MM, Lipowsky HH, Karp N, Schmalzer E, Chein S. A model of microvascular oxygen transport in sickle cell disease. MicrovascRes 1985; 30:195-206.
