Original Contributions
A Randomized ClinicalTrial of Rib Belts for SimpleFractures GARY QUICK, MD The authors pmsent a pilot study in which 20 patients with simple rib fractures were randomized prospectively into two treatment groups. One group mcelved ibuprofen and the other group lbupmfen plus a rib belt for analgesla. Them worn no statistically significant diffemnces obsenred in pulmonary function testing between the groups at initial visit, 48 hours, or 5 days. Atelectasis developed in four patients, two in each treatment gmup; them were no cases of pneumonltls. Patlents with displaced rib fmctums experienced a higher rata of bemo- or pneumothoraxthan did those with nondlsplaced fmctums (S/f0 v l/10). Patients wltb displaced fractures who used rib belts experienced a higher rate of hemothoraxthan those uslng oral analgesla alone (4/S v l/4). Patients uslng rib belts uniformly reported a significant amount of additional pain mlief. The clinician can use a rib belt to provide additional comfort to the patient with fractured ribs without appamnt addltlonal compromiseto respiratory pammeters. A furtfrer study stratifying displaced and nondisplacedfmcturns has been initiated to clarify the possible contributingroles of displaced rib fmctums and the rib belt In patients with displaced fractures. (Am J Emerg Med 1990;8:277-281. 0 1990 by W.8. SaundersCompany.)
Rib fractures are the most common sequel of blunt thoracic trauma. Currently most clinicians recommend oral analgesia for outpatient care of patients with uncomplicated rib fractures, ie, those without pneumothorax or hemothorax. Patients with simple rib fractures have been treated with a variety of oral analgesics, taping, or intercostal nerve blocks for pain relief. In 1945, rib belts became widely available and were used in the treatment of rib fractures, stabilizing the chest wall led to improved pain relief to the patient. Over the past 20 years the use of rib belts has fallen into disfavor because of the belief that use of a rib belt compromises chest wall expansion and underlying lung function that may lead to an increased frequency of atelectasis and secondary pneumonitis. ’ We could find no reliable data in the literature to support this opinion. Furthermore, the literature had nothing on the number of respiratory impairments caused by uncomplicated rib fractures.
From the Emergency Department, Barnes Hospital, and the Department of Emergency Medicine, Washington University School of Medicine. Manuscriot received ADrii 28. 1989; revision accepted August 3, 1989. ’ Address reprint requests to Dr Quick: Emergency Department, Barnes Hospital, One Barnes Hospital Plaza, St. Louis, MO 83110. ’ Key Words: Rib belt, rib fractures, outpatient management of rib fractures, hemothorax, possible complication of rib belt, effect on pulmonary function testing. 8 1990 by W.B. Saunders Company. 0735-8757/90/0804-0001$5.00/0
A pilot study was conducted on patients with uncomplicated rib fractures to describe and measure the alterations in pulmonary function testing and pulse oximetry noted in patients taking ibuprofen alone compared with those taking ibuprofen and wearing a rib belt for analgesia. The study was also intended to determine whether rib belt-treated patients experienced a higher number of complications, specifically atelectasis or pneumonitis, as compared with patients treated with oral analgesics alone. There is considerable evidence that prolonged restriction of respiratory excursion produces atelectasis which, in turn, leads to pneumonitis.’ The cycle of thoracic restriction leading to atelectasis and pneumonitis may be seen in other clinical settings such as restrictive lung disease, restrictive chest wall disease, and voluntary splinting due to pain. Personal clinical experience over the past 12 years using rib belts in patient case 10 to 12 times per year, particularly for those patients whose pain was inadequately relieved by oral analgesics, has not shown the use of rib belts to cause additional complications. Rather, the patients experienced an additional measure of pain relief quickly and without notable complications such as atelectasis or pneumonitis. MATERIALS AND METHODS Twenty consecutive patients with radiographically identified rib fractures, presenting to the emergency department at Scott and White Hospital during the period from October 1984 through October 1987, were randomized prospectively into two treatment groups by a computer-generated randomization schedule. The first group (OA) received oral analgesia in the form of ibuprofen 600 mg four times daily as sole treatment for rib fracture. The second group (RB) received the same ibuprofen regimen plus a Universal elastic and Velcro rib belt (Universal elastic rib belt, catalog #2722-M, Patient Care Systems, Zimmer, Warsaw, IN; Velcro, USA, Inc, Manchester, NH) applied initially with the top edge of the belt at the xiphoid process and applied tightly enough to provide optimal relief of pain to each patient. Data collected on initial, 48 hour, and 5 day follow-up visits included pulmonary function testing of forced expiratory volume in one second (FEVi), forced vital capacity (FVC), and negative inspiratory pressure (NIP). Other parameters collected on each visit were standard posteroanterior (PA) and lateral chest radiograph, and measurement of arterial oxygen saturation by pulse oximeter. Measurements on the initial visit were made before and after application of the rib belt. On each follow-up visit data were collected with 277
278
AMERICAN JOURNAL OF EMERGENCY MEDICINE W Volume 8, Number 4 W July 1990
or without rib belt in place, depending on whether the patient was wearing the belt at time of presentation. Pulmonary function was measured using a Breon portable spirometer (Breon spirometer model #2400, Breon, NY). The best of two patient efforts was recorded as is standard procedure for our pulmonary function testing for both FEV, and FVC. Pulse oximetry (PO) was measured by means of the Nellcor pulse oximeter (model N-lOOC, Nellcor, Inc, Hayward, CA). Each patient was allowed to wear the rib belt in a fashion giving optimal pain relief. The position of the belt was recorded at each follow-up visit. If a patient experienced inadequate pain relief from ibuprofen, therapy was switched to acetaminophen with 30 mg codeine. If a patient was admitted to the hospital for pain treatment during the study, the treatment protocol was continued. If chest tube placement was required, only the visits before tube thoracostomy were considered in data analysis. Definitive radiographic interpretation was provided by staff radiologists to determine the presence of atelectasis, pneumonitis, or other complications. It was the consensus of the staffradiologists that the 48-hour chest radiograph would identify the highest incidence of atelectasis and the S-day radiograph would respectively identify the greatest incidence of pneumonitis postinjury. All data were recorded on the study data sheet. Data analysis was accomplished with the assistance of the Biostatistics Department. Complication rates between the two study groups were compared using Fisher’s exact test. The change in pulmonary function and pulse oximetry variables from initial visit to both the 48-hour visit and the 5-day visit were compared using independent sample t-tests. All statistical comparisons are based on two-tailed tests, and P values of < .05 are considered significant. Our experimental design was reviewed and approved by the institution’s research committee. Informed consent was obtained for each participant and four nurses served as research assistants to collect and record data. Spirometric standards were obtained from the prediction nomogram of Morris et al* that incorporates age and height into the nomographic standard values.
RESULTS The study population comprised 15 men and five women averaging 48 years of age (Table 1). Eleven patients were randomized into treatment group RI3 and nine into group OA. The patients proved to have an average of 2.65 fractured ribs each. None had isolated upper rib fractures (ribs 1 through 4); 17 had varying combinations of middle rib fractures (ribs 5 through 9); three patients had lower fractures (ribs 10 through 12). The initial injury produced a 2% drop in FVC from the predicted value with decreases ranging from 60% to 10%. Nellcor pulse oximeter values showed arterial saturation of 97.8% for OA and %.8% for RB. Table 2 shows the mean and standard deviation for the pulmonary function and pulse oximetry variables by study group. The following complications were noted during the study (Table 3). Four patients developed atelectasis, two in each group. Of these four, two had displaced and two had nondisplaced fractures. No patient developed clinical or radiographic evidence of pneumonitis during the study. Three pa-
TABLE1.
Demographic Data
Total Patients
Rib Belt
No Belt
Sex
Male Female Trauma Fall MVA Other Smoker Yes No Diseases PUD COPD Asthma Diabetes Bronchitis Age (Y) Patients Mean Standard deviation Youngest Oldest
8 3
7 2
8
3 5 1
2 2 9 1 0 0 1 11
9
52 25 15 91
43 17 22 89
ABBREVIATIONS:
WA, motor vehicle accident; PUD, peptic ulcer disease; COPD, chronic obstructive pulmonary disease.
tients were noted to have small hemo- or pneumothorax that cleared without chest tube insertion. One patient in group OA experienced initial, small pneumothorax and developed a pulmonary infiltrate at 24 hours which cleared without antibiotic therapy. It was believed she had pulmonary contusion rather than pneumonitis, although she did have a febrile course for 48 hours. Two patients required tube thoracostomy, one for 50% pneumothorax at 48 hours and the other for expanding hemothorax at one month postinjury. Both had suffered multiple displaced rib fractures of the middle ribs. Two patients required hospitalization during the course of the study, one for pulmonary contusion and pain relief and the other for chest tube placement for 50% pneumothorax.
$6 >
Y’ u_ 0 -
I
-
BELT
---
ORAL
ANALGESIC
I
INITIAL
48 HR HOSPITAL VISITS
5 DAY
FIGURE 1. FVC values for all patients returning for all follow-up visits. (N = 20)
GARY QUICK W RIB BELTS FOR SIMPLE FRACTURES
TABLE2.
279
Pulmonary Function and Pulse Oximetry Variables (Mean f Standard Deviation) Initial RB
FVC FEV, NIP PO
2.9 2.1 -48.1 96.7
48 Hour RB
OA
_’ 1.1 2 0.9 k 21.6 2 1.5
2.5 1.9 - 52.0 97.8
+ + f 2
0.9 0.7 17.5 1.5
3.0 2.3 -52.8 96.5
2 + + k
5 Day
OA
0.8 0.6 16.0 1.6
3.0 2.3 - 68.9 97.7
2 + + 2
P Value*
0.8 0.9 23.0 1.8
0.17 0.30 0.26 0.84
RB
3.3 2.6 -49.5 97.5
+ 1.2 f 1.1 f 18.4 It_0.9
OA
3.8 2.9 -58.0 93.2
+ z k +
P
0.8 0.7 12.7 9.9
Value’
0.11 0.84 0.97 0.12
’ P values for comparing change from Initial between RB and OA.
pneumonitis, we observed equal incidence of atelectasis and our study groups and no pneumonitis in either group. We did, however, Observe an apparent increased incidence of small hemothorax in the RB study group, a finding previously unreported. The following studies help to establish the setting for our study. For-man performed a small uncontrolled study using rib belts on normal subjects of varying age and weight in order to determine the maximal degree of chest wall restriction. He found that the most significant pulmonary measurement was FVC and that after application of the belt, the FVC ranged between 70% to 80% of normal. He provided no further data in support of his conclusions.4 Pedersen et al5 measured maximum respiratory flow rates in 10 patients with rib fractures both before and after analgesia with intercostal nerve block. Two significant conclusions came from the study: (1) the preblock forced expiratory flow rates varied over a wide range and were unrelated to the number of injured ribs (one to four); (2) respiratory function in all patients
No significant difference in pulmonary function testing was noted between the groups (Fig 1 and Table 2). Subjectively, patients felt that use of a rib belt contributed approximately 25% additional pain relief over oral analgesia alone. All patients in RB reported increased relief of pain while wearing the belt. Five wore their belt at the xiphoid, three preferred to place the top of the belt 10 cm to 12 cm above the xiphoid, and three preferred 1 cm to 3 cm below the xiphoid. Most reported wearing their rib belt between 3 and 7 days with some as long as 1 month. DISCUSSION The central fact discovered in our study is that there appears to be no significant difference in pulmonary function testing or in incidence of atelectasis or pneumonitis as complications noted with use of rib belts and oral analgesia when compared with oral analgesia alone. Contrary to the commonly embraced concept that use of rib belts is associated with an increased incidence of atelectasis and secondary TABLE3.
Pt
FVC and Complications
Sorted by Study Group
Displaced (D)/ Nondisplaced (N)
Gender
Age
Group OA 1 Male 2 Male 3 Female
27 38 69
D N D
FVC in Liters 48 Hr
5 Day
Complications
5.4 5.3 3.8
2.6 4.0 1.0
4.3 4.3 1.0
4.7 3.5
None None Pneumothorax, pulmonary contusion None Atelectasis Hemothorax, chest tube (1 mo) None Atelectasis None
8 5-7 2-7
Hemothorax, None None None None None Nonethorax Hemothorax Atelectasis, hemo-, pneumothorax Pneumothorax, chest tube Atelectasis, hemothorax
8-9
4 5 8
Female Male Male
60 53 54
D N D
3.1 3.3 3.7
2.2 2.2 1.5
1.8 2.9 3.6
7 8 9
Male Male Male
22 37 24
N N
4.2 3.7 5.8
2.8 3.1 3.4
3.0
10 11 12 13 14 15 16 17 18
Female Male Male Male Male Male Male Female Male
61 32 40 90 28 28 16 67 69
3.5 5.2 5.4 4.0 5.5 5.8 5.8 2.8 2.8
3.1 3.0 5.2 2.8
3.3 3.3 4.0 3.0 3.0
3.7 3.5 2.2 2.5
3.5 2.3 2.5
19
Male
63
D
3.3
1.8
2.3
20
Female
79
D
2.5
Group
D
Fractured Ribs (No.)
Initial
Predict
2.5 3.4 3.5 4.2
3.2
8 9-11 4-9 1 5-7 4.9
RB
1.6
3.3 5.0 2.3 3.0 5.4 3.6 2.4 3.1
1.6
6 7-9 11-12 6 10 8-7 2-7 3-8 8-l 1
280
AMERICANJOURNAL OF EMERGENCY MEDICINEH Volume 8, Number 4 m July 1990
had improved 1 hour postblockade (P < 9001). Retesting results at 6 hours postblockade showed values to have returned to preblockade levels. Kappey and Mulled evaluated patients with severe blunt chest injuries to determine the functional background and the extent of respiratory insufficiency using daily arterial blood gas analysis (ABG) during the first week after trauma in 15 cases of unilateral serial rib fractures. They noted initial uncompensated metabolic acidosis on day 1, progressing to a temporary respiratory acidosis with metabolic compensation over a few days. By 1 week hypercapnia resolved and hypoxemia improved significantly. Another 21 patients with the same type of injury were assessed with spirometry. Starting on the fourth day postinjury, because of inability to obtain patient cooperation on day 1 through 3, nine estimations were done on each patient within 12 months. Vital capacity, total capacity, and maximum breathing capacity were markedly diminished initially, increasing sharply shortly after injury and reaching the normal range in about 6 weeks. The functional disturbances thus appeared to be of purely restrictive origin caused by the impairment of chest wall movements due to posttraumatic pain and thoracic instability.6 Many experts recommend hospitalization for patients experiencing greater than two rib fractures, advanced age, jagged rib fragments with displacement, and inability to provide adequate analgesia.4 Ten of our patients had displaced fractures and nine had greater than two ribs fractured. Only two patients were hospitalized acutely and one patient was hospitalized 1 month postinjury. The small sample size is a limitation that must be taken into consideration in the interpretation of the results. Although there was no difference between groups in this study, the possibility of a type II error must be considered, ie, the possibility that the treatments make a difference in the risk of atelectasis or pneumonitis variation. The power of the experiment is less than 50% for detecting a difference between groups as large as 10% atelectasis versus 30% atelectasis (p > 0.5). In order to plan an experiment with adequate power (90% power or l3 = 0.1) for rates of 10% versus 30%, a sample size of 164 patients would be needed (82 in each group). This sample size would be needed to minimize the chance of missing a difference as large as 10% versus 30%. The smaller the difference one would wish to detect, the larger the required sample size. In order to have 90% power for a difference as small as 15% versus 25%, it would be necessary to recruit 668 patients (334 in each group). It is clear from the distribution of complications noted in the study that displaced rib fractures carry a higher rate of complications. Hemothorax or pneumothorax occurred in 50% of our patients with displaced fracture (Table 4). We note with interest the occurrence of possible increased incidence of hemothorax in the RB study group. This complication developed in patients who also had experienced multiple displaced rib fractures. We are unable to state with con& dence whether this finding of hemothorax represents a coincidental finding because of small sample size, or whether there is truly a correlation between multiple displaced rib fractures and the use of rib belts, acting as possible cofactors, contributing to hemothorax as a complication. Until
TABLE4. Proportion of Patients With Pneumothorax Hemothorax Within Five Days of Injury Complication No displacement (l-3 ribs fractured) Displacement (1-3 ribs fractured) Displacement (4-6 ribs fractured)
and/or
OA
RB
Total
o/5
l/5
l/10
O/l
l/3
l/4
l/3
313
416
this area is further clarified by a stratified study, we recommend caution with the use of rib belts in patients with multiple displaced fractures. Because of the severity of injury in our patient population, we believe that a 20% incidence of atelectasis represents a generous estimate of the incidence of atelectasis occurring in patients with rib fractures. In Dougherty’s’ study, for instance, the patient population was limited to those patients having one or two nondisplaced fractures. Many of the patients in our study had two to six displaced or nondisplaced fractures. Therefore, we can say that the frequency of atelectasis probably does not exceed 20% in most patients with rib fractures believed suitable for outpatient care and that pneumonitis is probably not encountered any more frequently as a complication in patients with the more routine one to two rib fractures than in our population. Dougherty’ reported two cases of asymptomatic atelectasis at the 2-week postinjury interval and also noted one case of dermatitis thought to be related to use of a rib belt in his patient population of 40 patients, 15 of whom wore the rib belts. We did not evaluate for atelectasis at that interval because our radiology consultants recommended the 48-hour interval as the most likely to demonstrate atelectasis and the live-day interval for pneumonitis. We had no patients who developed contact dermatitis in our study. Dougherty also reported one case of bloody pleural effusion requiring hospitalization, but because of other underlying problems, ie, alcoholism, it is unclear whether the rib belt was the sole cause of the effusion. pulse oximetry, an effective noninvasive monitoring technique, proved to be a reliable means of monitoring arterial saturation in patients with rib fractures. The Nellcor pulse oximeter has a saturation range of 0% to 100% and an accuracy of +2 digits over a range of 70% to 100% arterial saturation. We think that the trend of pulse oximetry measurement of oxygen saturation can be a reliable, noninvasive, conveniently obtained indicator of clinically significant arterial desaturation in patients with rib fractures. No patients demonstrated significant desaturation or hypoxemia based either on ABG analysis or on pulse oximetry in this study. We conclude that most patients with simple rib fractures can be treated by oral analgesics or by rib belt without apparent additional compromise to pulmonary mechanics or risk of atelectasis or pneumonitis with either treatment modality. A further study stratifying displaced and nondisplaced fractures has been initiated to clarify the possible contributory roles of displaced rib fractures and use a rib belt in patients with displaced fractures who experience hemo- or pneumothorax.
GARY QUICK n RIB BELTS FOR SIMPLE FRACTURES
1. Rutherford RF: Thoracic injuries, in Ballinger WF, Rutherford RF, Zuidema GA (eds): The Management of Trauma, ed 2. Philadelphia, PA, Saunders 1973, pp. 344-345 2. Morris JF, Koski A, Johnson LC: Spirometric standards for healthy nonsmoking adults. Am Rev Resp Dis 1971;103:57-67 3. Pennock BE, Cottrell JJ, Rogers RM: Pulmonary function testing: What is “normal”? Arch Intern Med 1983;143:2123-2127
281
4. Forman EW: Rib belts for chest wall pain. Aches Pains 1983;10:20-25 5. Pedersen VM, Schulze S, Madsen K, Halkier E: Air-flow meter assessment of the effect of intercostal nerve blockade on respiratory function in rib fractures. Acta Chir Stand 1983;149: 119-120 6. Kappey F, Muller C: Respiratory problems following chest trauma. Excerpta Medica Surg 1971;25:466 7. Dougherty J: Interview with Dougherty J. Emerg Med Ambulatory Care News 1987;9:5
A Randomized ClinicalTrial of Rib Belts for SimpleFractures GARY QUICK, MD The authors pmsent a pilot study in which 20 patients with simple rib fractures were randomized prospectively into two treatment groups. One group mcelved ibuprofen and the other group lbupmfen plus a rib belt for analgesla. Them worn no statistically significant diffemnces obsenred in pulmonary function testing between the groups at initial visit, 48 hours, or 5 days. Atelectasis developed in four patients, two in each treatment gmup; them were no cases of pneumonltls. Patlents with displaced rib fmctums experienced a higher rata of bemo- or pneumothoraxthan did those with nondlsplaced fmctums (S/f0 v l/10). Patients wltb displaced fractures who used rib belts experienced a higher rate of hemothoraxthan those uslng oral analgesla alone (4/S v l/4). Patients uslng rib belts uniformly reported a significant amount of additional pain mlief. The clinician can use a rib belt to provide additional comfort to the patient with fractured ribs without appamnt addltlonal compromiseto respiratory pammeters. A furtfrer study stratifying displaced and nondisplacedfmcturns has been initiated to clarify the possible contributingroles of displaced rib fmctums and the rib belt In patients with displaced fractures. (Am J Emerg Med 1990;8:277-281. 0 1990 by W.8. SaundersCompany.)
Rib fractures are the most common sequel of blunt thoracic trauma. Currently most clinicians recommend oral analgesia for outpatient care of patients with uncomplicated rib fractures, ie, those without pneumothorax or hemothorax. Patients with simple rib fractures have been treated with a variety of oral analgesics, taping, or intercostal nerve blocks for pain relief. In 1945, rib belts became widely available and were used in the treatment of rib fractures, stabilizing the chest wall led to improved pain relief to the patient. Over the past 20 years the use of rib belts has fallen into disfavor because of the belief that use of a rib belt compromises chest wall expansion and underlying lung function that may lead to an increased frequency of atelectasis and secondary pneumonitis. ’ We could find no reliable data in the literature to support this opinion. Furthermore, the literature had nothing on the number of respiratory impairments caused by uncomplicated rib fractures.
From the Emergency Department, Barnes Hospital, and the Department of Emergency Medicine, Washington University School of Medicine. Manuscriot received ADrii 28. 1989; revision accepted August 3, 1989. ’ Address reprint requests to Dr Quick: Emergency Department, Barnes Hospital, One Barnes Hospital Plaza, St. Louis, MO 83110. ’ Key Words: Rib belt, rib fractures, outpatient management of rib fractures, hemothorax, possible complication of rib belt, effect on pulmonary function testing. 8 1990 by W.B. Saunders Company. 0735-8757/90/0804-0001$5.00/0
A pilot study was conducted on patients with uncomplicated rib fractures to describe and measure the alterations in pulmonary function testing and pulse oximetry noted in patients taking ibuprofen alone compared with those taking ibuprofen and wearing a rib belt for analgesia. The study was also intended to determine whether rib belt-treated patients experienced a higher number of complications, specifically atelectasis or pneumonitis, as compared with patients treated with oral analgesics alone. There is considerable evidence that prolonged restriction of respiratory excursion produces atelectasis which, in turn, leads to pneumonitis.’ The cycle of thoracic restriction leading to atelectasis and pneumonitis may be seen in other clinical settings such as restrictive lung disease, restrictive chest wall disease, and voluntary splinting due to pain. Personal clinical experience over the past 12 years using rib belts in patient case 10 to 12 times per year, particularly for those patients whose pain was inadequately relieved by oral analgesics, has not shown the use of rib belts to cause additional complications. Rather, the patients experienced an additional measure of pain relief quickly and without notable complications such as atelectasis or pneumonitis. MATERIALS AND METHODS Twenty consecutive patients with radiographically identified rib fractures, presenting to the emergency department at Scott and White Hospital during the period from October 1984 through October 1987, were randomized prospectively into two treatment groups by a computer-generated randomization schedule. The first group (OA) received oral analgesia in the form of ibuprofen 600 mg four times daily as sole treatment for rib fracture. The second group (RB) received the same ibuprofen regimen plus a Universal elastic and Velcro rib belt (Universal elastic rib belt, catalog #2722-M, Patient Care Systems, Zimmer, Warsaw, IN; Velcro, USA, Inc, Manchester, NH) applied initially with the top edge of the belt at the xiphoid process and applied tightly enough to provide optimal relief of pain to each patient. Data collected on initial, 48 hour, and 5 day follow-up visits included pulmonary function testing of forced expiratory volume in one second (FEVi), forced vital capacity (FVC), and negative inspiratory pressure (NIP). Other parameters collected on each visit were standard posteroanterior (PA) and lateral chest radiograph, and measurement of arterial oxygen saturation by pulse oximeter. Measurements on the initial visit were made before and after application of the rib belt. On each follow-up visit data were collected with 277
278
AMERICAN JOURNAL OF EMERGENCY MEDICINE W Volume 8, Number 4 W July 1990
or without rib belt in place, depending on whether the patient was wearing the belt at time of presentation. Pulmonary function was measured using a Breon portable spirometer (Breon spirometer model #2400, Breon, NY). The best of two patient efforts was recorded as is standard procedure for our pulmonary function testing for both FEV, and FVC. Pulse oximetry (PO) was measured by means of the Nellcor pulse oximeter (model N-lOOC, Nellcor, Inc, Hayward, CA). Each patient was allowed to wear the rib belt in a fashion giving optimal pain relief. The position of the belt was recorded at each follow-up visit. If a patient experienced inadequate pain relief from ibuprofen, therapy was switched to acetaminophen with 30 mg codeine. If a patient was admitted to the hospital for pain treatment during the study, the treatment protocol was continued. If chest tube placement was required, only the visits before tube thoracostomy were considered in data analysis. Definitive radiographic interpretation was provided by staff radiologists to determine the presence of atelectasis, pneumonitis, or other complications. It was the consensus of the staffradiologists that the 48-hour chest radiograph would identify the highest incidence of atelectasis and the S-day radiograph would respectively identify the greatest incidence of pneumonitis postinjury. All data were recorded on the study data sheet. Data analysis was accomplished with the assistance of the Biostatistics Department. Complication rates between the two study groups were compared using Fisher’s exact test. The change in pulmonary function and pulse oximetry variables from initial visit to both the 48-hour visit and the 5-day visit were compared using independent sample t-tests. All statistical comparisons are based on two-tailed tests, and P values of < .05 are considered significant. Our experimental design was reviewed and approved by the institution’s research committee. Informed consent was obtained for each participant and four nurses served as research assistants to collect and record data. Spirometric standards were obtained from the prediction nomogram of Morris et al* that incorporates age and height into the nomographic standard values.
RESULTS The study population comprised 15 men and five women averaging 48 years of age (Table 1). Eleven patients were randomized into treatment group RI3 and nine into group OA. The patients proved to have an average of 2.65 fractured ribs each. None had isolated upper rib fractures (ribs 1 through 4); 17 had varying combinations of middle rib fractures (ribs 5 through 9); three patients had lower fractures (ribs 10 through 12). The initial injury produced a 2% drop in FVC from the predicted value with decreases ranging from 60% to 10%. Nellcor pulse oximeter values showed arterial saturation of 97.8% for OA and %.8% for RB. Table 2 shows the mean and standard deviation for the pulmonary function and pulse oximetry variables by study group. The following complications were noted during the study (Table 3). Four patients developed atelectasis, two in each group. Of these four, two had displaced and two had nondisplaced fractures. No patient developed clinical or radiographic evidence of pneumonitis during the study. Three pa-
TABLE1.
Demographic Data
Total Patients
Rib Belt
No Belt
Sex
Male Female Trauma Fall MVA Other Smoker Yes No Diseases PUD COPD Asthma Diabetes Bronchitis Age (Y) Patients Mean Standard deviation Youngest Oldest
8 3
7 2
8
3 5 1
2 2 9 1 0 0 1 11
9
52 25 15 91
43 17 22 89
ABBREVIATIONS:
WA, motor vehicle accident; PUD, peptic ulcer disease; COPD, chronic obstructive pulmonary disease.
tients were noted to have small hemo- or pneumothorax that cleared without chest tube insertion. One patient in group OA experienced initial, small pneumothorax and developed a pulmonary infiltrate at 24 hours which cleared without antibiotic therapy. It was believed she had pulmonary contusion rather than pneumonitis, although she did have a febrile course for 48 hours. Two patients required tube thoracostomy, one for 50% pneumothorax at 48 hours and the other for expanding hemothorax at one month postinjury. Both had suffered multiple displaced rib fractures of the middle ribs. Two patients required hospitalization during the course of the study, one for pulmonary contusion and pain relief and the other for chest tube placement for 50% pneumothorax.
$6 >
Y’ u_ 0 -
I
-
BELT
---
ORAL
ANALGESIC
I
INITIAL
48 HR HOSPITAL VISITS
5 DAY
FIGURE 1. FVC values for all patients returning for all follow-up visits. (N = 20)
GARY QUICK W RIB BELTS FOR SIMPLE FRACTURES
TABLE2.
279
Pulmonary Function and Pulse Oximetry Variables (Mean f Standard Deviation) Initial RB
FVC FEV, NIP PO
2.9 2.1 -48.1 96.7
48 Hour RB
OA
_’ 1.1 2 0.9 k 21.6 2 1.5
2.5 1.9 - 52.0 97.8
+ + f 2
0.9 0.7 17.5 1.5
3.0 2.3 -52.8 96.5
2 + + k
5 Day
OA
0.8 0.6 16.0 1.6
3.0 2.3 - 68.9 97.7
2 + + 2
P Value*
0.8 0.9 23.0 1.8
0.17 0.30 0.26 0.84
RB
3.3 2.6 -49.5 97.5
+ 1.2 f 1.1 f 18.4 It_0.9
OA
3.8 2.9 -58.0 93.2
+ z k +
P
0.8 0.7 12.7 9.9
Value’
0.11 0.84 0.97 0.12
’ P values for comparing change from Initial between RB and OA.
pneumonitis, we observed equal incidence of atelectasis and our study groups and no pneumonitis in either group. We did, however, Observe an apparent increased incidence of small hemothorax in the RB study group, a finding previously unreported. The following studies help to establish the setting for our study. For-man performed a small uncontrolled study using rib belts on normal subjects of varying age and weight in order to determine the maximal degree of chest wall restriction. He found that the most significant pulmonary measurement was FVC and that after application of the belt, the FVC ranged between 70% to 80% of normal. He provided no further data in support of his conclusions.4 Pedersen et al5 measured maximum respiratory flow rates in 10 patients with rib fractures both before and after analgesia with intercostal nerve block. Two significant conclusions came from the study: (1) the preblock forced expiratory flow rates varied over a wide range and were unrelated to the number of injured ribs (one to four); (2) respiratory function in all patients
No significant difference in pulmonary function testing was noted between the groups (Fig 1 and Table 2). Subjectively, patients felt that use of a rib belt contributed approximately 25% additional pain relief over oral analgesia alone. All patients in RB reported increased relief of pain while wearing the belt. Five wore their belt at the xiphoid, three preferred to place the top of the belt 10 cm to 12 cm above the xiphoid, and three preferred 1 cm to 3 cm below the xiphoid. Most reported wearing their rib belt between 3 and 7 days with some as long as 1 month. DISCUSSION The central fact discovered in our study is that there appears to be no significant difference in pulmonary function testing or in incidence of atelectasis or pneumonitis as complications noted with use of rib belts and oral analgesia when compared with oral analgesia alone. Contrary to the commonly embraced concept that use of rib belts is associated with an increased incidence of atelectasis and secondary TABLE3.
Pt
FVC and Complications
Sorted by Study Group
Displaced (D)/ Nondisplaced (N)
Gender
Age
Group OA 1 Male 2 Male 3 Female
27 38 69
D N D
FVC in Liters 48 Hr
5 Day
Complications
5.4 5.3 3.8
2.6 4.0 1.0
4.3 4.3 1.0
4.7 3.5
None None Pneumothorax, pulmonary contusion None Atelectasis Hemothorax, chest tube (1 mo) None Atelectasis None
8 5-7 2-7
Hemothorax, None None None None None Nonethorax Hemothorax Atelectasis, hemo-, pneumothorax Pneumothorax, chest tube Atelectasis, hemothorax
8-9
4 5 8
Female Male Male
60 53 54
D N D
3.1 3.3 3.7
2.2 2.2 1.5
1.8 2.9 3.6
7 8 9
Male Male Male
22 37 24
N N
4.2 3.7 5.8
2.8 3.1 3.4
3.0
10 11 12 13 14 15 16 17 18
Female Male Male Male Male Male Male Female Male
61 32 40 90 28 28 16 67 69
3.5 5.2 5.4 4.0 5.5 5.8 5.8 2.8 2.8
3.1 3.0 5.2 2.8
3.3 3.3 4.0 3.0 3.0
3.7 3.5 2.2 2.5
3.5 2.3 2.5
19
Male
63
D
3.3
1.8
2.3
20
Female
79
D
2.5
Group
D
Fractured Ribs (No.)
Initial
Predict
2.5 3.4 3.5 4.2
3.2
8 9-11 4-9 1 5-7 4.9
RB
1.6
3.3 5.0 2.3 3.0 5.4 3.6 2.4 3.1
1.6
6 7-9 11-12 6 10 8-7 2-7 3-8 8-l 1
280
AMERICANJOURNAL OF EMERGENCY MEDICINEH Volume 8, Number 4 m July 1990
had improved 1 hour postblockade (P < 9001). Retesting results at 6 hours postblockade showed values to have returned to preblockade levels. Kappey and Mulled evaluated patients with severe blunt chest injuries to determine the functional background and the extent of respiratory insufficiency using daily arterial blood gas analysis (ABG) during the first week after trauma in 15 cases of unilateral serial rib fractures. They noted initial uncompensated metabolic acidosis on day 1, progressing to a temporary respiratory acidosis with metabolic compensation over a few days. By 1 week hypercapnia resolved and hypoxemia improved significantly. Another 21 patients with the same type of injury were assessed with spirometry. Starting on the fourth day postinjury, because of inability to obtain patient cooperation on day 1 through 3, nine estimations were done on each patient within 12 months. Vital capacity, total capacity, and maximum breathing capacity were markedly diminished initially, increasing sharply shortly after injury and reaching the normal range in about 6 weeks. The functional disturbances thus appeared to be of purely restrictive origin caused by the impairment of chest wall movements due to posttraumatic pain and thoracic instability.6 Many experts recommend hospitalization for patients experiencing greater than two rib fractures, advanced age, jagged rib fragments with displacement, and inability to provide adequate analgesia.4 Ten of our patients had displaced fractures and nine had greater than two ribs fractured. Only two patients were hospitalized acutely and one patient was hospitalized 1 month postinjury. The small sample size is a limitation that must be taken into consideration in the interpretation of the results. Although there was no difference between groups in this study, the possibility of a type II error must be considered, ie, the possibility that the treatments make a difference in the risk of atelectasis or pneumonitis variation. The power of the experiment is less than 50% for detecting a difference between groups as large as 10% atelectasis versus 30% atelectasis (p > 0.5). In order to plan an experiment with adequate power (90% power or l3 = 0.1) for rates of 10% versus 30%, a sample size of 164 patients would be needed (82 in each group). This sample size would be needed to minimize the chance of missing a difference as large as 10% versus 30%. The smaller the difference one would wish to detect, the larger the required sample size. In order to have 90% power for a difference as small as 15% versus 25%, it would be necessary to recruit 668 patients (334 in each group). It is clear from the distribution of complications noted in the study that displaced rib fractures carry a higher rate of complications. Hemothorax or pneumothorax occurred in 50% of our patients with displaced fracture (Table 4). We note with interest the occurrence of possible increased incidence of hemothorax in the RB study group. This complication developed in patients who also had experienced multiple displaced rib fractures. We are unable to state with con& dence whether this finding of hemothorax represents a coincidental finding because of small sample size, or whether there is truly a correlation between multiple displaced rib fractures and the use of rib belts, acting as possible cofactors, contributing to hemothorax as a complication. Until
TABLE4. Proportion of Patients With Pneumothorax Hemothorax Within Five Days of Injury Complication No displacement (l-3 ribs fractured) Displacement (1-3 ribs fractured) Displacement (4-6 ribs fractured)
and/or
OA
RB
Total
o/5
l/5
l/10
O/l
l/3
l/4
l/3
313
416
this area is further clarified by a stratified study, we recommend caution with the use of rib belts in patients with multiple displaced fractures. Because of the severity of injury in our patient population, we believe that a 20% incidence of atelectasis represents a generous estimate of the incidence of atelectasis occurring in patients with rib fractures. In Dougherty’s’ study, for instance, the patient population was limited to those patients having one or two nondisplaced fractures. Many of the patients in our study had two to six displaced or nondisplaced fractures. Therefore, we can say that the frequency of atelectasis probably does not exceed 20% in most patients with rib fractures believed suitable for outpatient care and that pneumonitis is probably not encountered any more frequently as a complication in patients with the more routine one to two rib fractures than in our population. Dougherty’ reported two cases of asymptomatic atelectasis at the 2-week postinjury interval and also noted one case of dermatitis thought to be related to use of a rib belt in his patient population of 40 patients, 15 of whom wore the rib belts. We did not evaluate for atelectasis at that interval because our radiology consultants recommended the 48-hour interval as the most likely to demonstrate atelectasis and the live-day interval for pneumonitis. We had no patients who developed contact dermatitis in our study. Dougherty also reported one case of bloody pleural effusion requiring hospitalization, but because of other underlying problems, ie, alcoholism, it is unclear whether the rib belt was the sole cause of the effusion. pulse oximetry, an effective noninvasive monitoring technique, proved to be a reliable means of monitoring arterial saturation in patients with rib fractures. The Nellcor pulse oximeter has a saturation range of 0% to 100% and an accuracy of +2 digits over a range of 70% to 100% arterial saturation. We think that the trend of pulse oximetry measurement of oxygen saturation can be a reliable, noninvasive, conveniently obtained indicator of clinically significant arterial desaturation in patients with rib fractures. No patients demonstrated significant desaturation or hypoxemia based either on ABG analysis or on pulse oximetry in this study. We conclude that most patients with simple rib fractures can be treated by oral analgesics or by rib belt without apparent additional compromise to pulmonary mechanics or risk of atelectasis or pneumonitis with either treatment modality. A further study stratifying displaced and nondisplaced fractures has been initiated to clarify the possible contributory roles of displaced rib fractures and use a rib belt in patients with displaced fractures who experience hemo- or pneumothorax.
GARY QUICK n RIB BELTS FOR SIMPLE FRACTURES
1. Rutherford RF: Thoracic injuries, in Ballinger WF, Rutherford RF, Zuidema GA (eds): The Management of Trauma, ed 2. Philadelphia, PA, Saunders 1973, pp. 344-345 2. Morris JF, Koski A, Johnson LC: Spirometric standards for healthy nonsmoking adults. Am Rev Resp Dis 1971;103:57-67 3. Pennock BE, Cottrell JJ, Rogers RM: Pulmonary function testing: What is “normal”? Arch Intern Med 1983;143:2123-2127
281
4. Forman EW: Rib belts for chest wall pain. Aches Pains 1983;10:20-25 5. Pedersen VM, Schulze S, Madsen K, Halkier E: Air-flow meter assessment of the effect of intercostal nerve blockade on respiratory function in rib fractures. Acta Chir Stand 1983;149: 119-120 6. Kappey F, Muller C: Respiratory problems following chest trauma. Excerpta Medica Surg 1971;25:466 7. Dougherty J: Interview with Dougherty J. Emerg Med Ambulatory Care News 1987;9:5
