Triage of Patients With Rib Fractures

Introduction

Rib fractures occur in more than 10% of all trauma patients. ^{1 -3} Mechanisms of injury range from high-impact chest wall trauma to low-energy ground-level falls in the elderly. Fractured ribs can be seen in the setting of severe polytrauma or as isolated injuries, and even a low number of involved ribs can cause severe acute and chronic pain in the young and considerable morbidity and mortality in the elderly. ^{4 -7}

Numerous studies have attempted to identify high-risk patients that are likely to experience serious complications. ^{4,5,8 -10} Advanced age, increasing number of fractured ribs, chest wall instability, and baseline nontraumatic comorbidities are among frequently reported factors used in rib fracture scoring and triage systems. ^{8,10 -12} These high-risk patients may benefit from intensive care unit (ICU) admission ^13,14 to allow for effective pain management, aggressive pulmonary hygiene, and close monitoring for respiratory complications. However, overly sensitive ICU admission criteria will result in overutilization of resources and may unnecessarily expose patients to iatrogenic complications.

At our level 1 trauma center, we had used a highly sensitive triage guideline for patients with blunt chest wall trauma that was primarily based on patient age and number of fractured ribs, resulting in ICU admission rates of up to 75% of patients with rib fractures. Many of these patients did well clinically and were transferred to a general floor within 48 hours of admission. We therefore implemented a revised triage guideline that put less emphasis on patients’ age and the number of fractured ribs to determine disposition in those with good functional status. We hypothesized that we could lower ICU admission rates without compromising patient safety.

Methods

The Institutional Review Board at Virginia Commonwealth University, Richmond, VA, USA, reviewed and approved this study. Included were adult blunt trauma patients with at least 1 rib fracture admitted following our revised rib fracture triage guideline from July 2018 through March 2019 and our original triage guideline from January 2017 through June 2017. Patients intubated at or upon arrival and those with concomitant severe traumatic brain injury or polytrauma mandating ICU admission regardless of rib fractures were excluded due to this study’s focus on appropriate disposition. Additionally, pregnant patients, prisoners, and those under age 18 years were excluded.

The original and revised triage guidelines are shown in Table 1. Age 65 years and older, and 3 or more fractured ribs were strict ICU admission criteria in the original guideline. The revised guideline allowed for more liberal admission to a general floor or stepdown bed in patients doing well clinically despite these risk factors. There was no significant change in treatment strategies for patients with rib fractures during the study period. Those admitted to the ICU underwent continuous pulse oximetry monitoring (SpO₂), frequent pain score assessment by nursing, pulmonary hygiene with respiratory therapy, and early mobilization. A multimodal approach to pain control with patient-controlled analgesia (PCA), oral narcotic and non-narcotic pain medication, topical anesthetic patches, and muscle relaxing medication was employed. In cases of difficult pain control, the anesthesia pain service was consulted for placement of an epidural catheter, or a pump for continuous infusion of local anesthetics (OnQ pump) was placed. Patients admitted to the floor primarily relied on spot check SpO₂ measurements, nursing-assisted incentive spirometer (IS) use, and pain control with oral medications, topical anesthetic patches, and PCA.

Table 1

Original and Revised Rib Fracture Triage Guidelines.

Abbreviation: ICU, intensive care unit.

Details on patient characteristics, injury details, outcomes, and complications were obtained from the trauma registry and electronic medical records, as applicable. Categorical data were described as number and percentage (N, %) and compared using Chi-squared and Fisher’s exact test, as appropriate. Continuous data were expressed as median (25th/75th percentile) and compared using the Wilcoxon-Mann-Whitney test. Univariate followed by multivariate analyses were used to determine independent predictors of complications and prolonged hospital stay. All risk factors associated with these outcomes in the univariate model with P ≤ .2 were entered into the multivariate model, in which P ≤ .05 was considered significant. Statistical analyses were performed using JMP Pro 14.3.0. 2018 SAS Institute Inc., Cary, NC.

Results

A total of 248 patients were admitted to the trauma service over 9 months in 2018 and 2019 after the institution of the revised rib fracture triage guideline. Throughout 6 months in 2017 following the original guideline, 207 patients were admitted. The ICU admission rate was significantly lower after the institution of the revised triage guideline at 73.43% versus 62.50% (P = .02). Under the original guideline, 24.67% (N = 37) of patients had been discharged from the ICU within 48 hours. This proportion significantly decreased to 14.84% (N = 23, P = .03) following the new guideline. Those admitted following the revised triage guideline were older and had more severe associated pulmonary injury but did not differ in the number of fractured ribs and baseline comorbidities (Table 2).

Table 2

Patient Characteristics and Injury Details Under the Original and Revised Triage Guidelines.

Abbreviation: IS, incentive spirometer.

^aStatistically significant.

^bChronic obstructive pulmonary disease, pulmonary fibrosis, or severe asthma.

^cCirrhosis, heart failure, or taking therapy for anticoagulation.

^dFlail segment, bilateral rib fractures, sternal or scapular fracture.

^eTachypnea (respiratory rate 20 per minute or higher) or hypoxia with need for oxygen supplementation.

^fAbbreviated Injury Scale abdomen ≥2.

^gPercentage of those with available data.

There was no statistically significant difference in respiratory complications, defined as the need for intubation, chest tube placement, or other invasive thoracic procedure, or development of pneumonia between the groups. Furthermore, the unplanned ICU admission rate and overall mortality were not significantly different between the 2 groups. An adverse composite outcome of respiratory complications, unplanned ICU transfer, difficult pain control (need for pain service consultation and placement of epidural or local anesthetic catheter), or death demonstrated no significant difference between the 2 groups (Table 3).

Table 3

Outcomes and Complications Under the Original and Revised Triage Guidelines.

Abbreviations: ICU, intensive care unit; LOS, length of stay.

^aStatistically significant.

^bOf those not initially admitted to the ICU.

^cNeed for pain service consult and placement of epidural or local anesthetic catheter.

^dIntubation, pneumonia, chest tube placement, or other invasive thoracic procedure.

^eUnplanned ICU transfer, difficult pain control, respiratory complication, or death.

All 4 patients who experienced respiratory complications admitted under the original triage guideline required intubation for respiratory distress and had been admitted to the ICU upon patient arrival. Seven patients experienced respiratory complications following the new triage guideline, 6 of whom had been admitted to the ICU upon patient arrival. One patient who had been admitted to the floor initially required ICU admission for dyspnea and underwent video-assisted thoracoscopy for a retained hemothorax, eventually recovering well. The other 4 unplanned ICU transfers in this patient group were not due to rib fracture-related complications. Median ICU length of stay (LOS) and total hospital LOS were longer by 1 day under the revised triage guideline at a median of 4 (4/6) versus 3 (3/6) days, P = .09, and 6 (4/10) versus 5 (3/9) days, P = .02, respectively.

In a univariate analysis including all 455 patients of both groups, advanced age, 3 or more rib fractures, unstable chest wall, hemothorax or pneumothorax or pulmonary contusions, concomitant abdominal injuries, baseline comorbidities, dyspnea upon admission, initial IS effort 750 mL or less, and severe pain upon admission, defined as 6 or higher on a 10-point scale, were associated with the adverse composite outcome and with a prolonged hospital stay of more than the median of 6 days. In the multivariate model, only initial IS effort less than 750 mL and dyspnea in the trauma bay remained associated with the adverse composite outcome with odds ratio (OR) and 95% CI of OR = 5.02 (95% CI 1.62-15.54, P < .01) and OR = 4.36 (95% CI 1.41-13.50, P = .01), respectively. Dyspnea in the trauma bay, 3 or more fractured ribs, initial IS effort less than 750 mL, and severe pain upon admission remained associated with prolonged hospital stay in a multivariate model with OR = 3.49 (95% CI 1.54-7.91, P < .01), OR = 2.78 (95% CI 1.10-7.03, P = .03), OR = 2.52 (95% CI 1.09-5.84, P = .03), and OR = 2.30 (95% CI 1.03-5.13, P = .04), respectively.

Discussion

Our revised rib fracture triage guideline put less emphasis on age and number of fractured ribs, the traditional indicators of a high-risk patient population after blunt chest wall trauma. Instead, it allowed for clinical judgment to dictate admission disposition if patients demonstrated good functional status despite advanced age or multiple rib fractures. This revision resulted in a significantly decreased ICU admission rate. It particularly lowered the number of patients admitted to the ICU only to be transferred to the general floor within 48 hours of hospital admission, most likely indicating a patient population that had been admitted to the ICU because of their “numbers” rather than clinical necessity.

With the implementation of the revised triage guideline, we also began to routinely assess patients’ pain and inspiratory effort upon admission via quick bedside determination of pain score from 0 to 10 and the initial effort with IS. While these functional parameters were not part of the triage guideline, our analysis suggests that they might in fact be excellent predictors of complications and prolonged hospital stay and should be considered for this purpose. Existing retrospective studies on this topic suggest that increasing vital capacity in patients with rib fractures is associated with decreased risk for pulmonary complications, discharge to an extended care facility as opposed to home, and shorter hospital LOS. ^15,16

In our current analysis, there is a large proportion of missing data on these parameters, particularly for patients admitted in 2017 under the original guideline, when this was not a routine part of patient assessment. If confirmed in future studies, an abbreviated functional assessment upon admission may be a much more useful tool for risk stratification and disposition planning than “traditional” parameters in patients with blunt chest wall trauma. In fact, recent studies in patients with chest wall trauma suggest that physiologic parameters and patient reserve are more useful assessment tools than age groups and anatomical descriptions. ^17,18 This is also consistent with the more recent realization in geriatric trauma patients that functional status and physiologic reserve, that is, frailty or the lack thereof, are more useful and accurate descriptors of patient risk than many traditional parameters. ^19,20

The only adverse consequence of the revised triage guideline we found was a significantly increased ICU and hospital LOS, though by 1 day only. In trauma patients, LOS is a difficult outcome measure to interpret as numerous nonmedical factors such as homelessness, insurance status, substance abuse, and psychiatric comorbidities are quite prevalent and influence LOS. Additionally, significantly more patients admitted following the revised guideline had an unstable chest wall, hemothorax or pneumothorax or pulmonary contusions, or dyspnea, all markers of a more severely injured patient population and a possible explanation for an increased LOS.

Despite a substantial proportion of patients with multiple fractured ribs, sternal fractures, scapula fractures, and hemothorax or pneumothorax, the morbidity and mortality of our patient population were lower than traditionally reported ^{1,3 -5,14} but consistent with the results of other, primarily more recent studies that similarly demonstrated improved outcomes. ^2,9,18 These changes likely not only reflect the positive impact of advanced critical care and aggressive ICU admission guidelines but also the widespread use of chest CT scans that have much higher sensitivity for rib fractures than chest X-ray, thus increasing the number of patients with a low number of rib fractures. ² In our analysis, we excluded patients with severe associated polytrauma that would have mandated ICU admission for these injuries given the study focus on improving rib fracture triage and ICU admission guidelines. Therefore, our results are more reflective of the actual morbidity and mortality of fractured ribs alone rather than their contribution to morbidity and mortality in patients with blunt chest wall trauma as a marker of severe injury burden and massive polytrauma.

The exclusion of patients primarily admitted to the ICU for reasons other than rib fractures is also one of the limitations of this study as it may have introduced selection bias. It is impossible to completely separate these patient populations as rib fractures rarely occur in isolation, and ICU versus floor admission is sometimes due to clinical gestalt, physician discretion, or even nonmedical, logistical reasons. Missing data and the general limitations of any retrospective review are additional considerations. Lastly, it is easy to underestimate the potential benefits of ICU care even in noncritical patients with rib fractures, that is, higher nurse-to-patient ratios, more intense respiratory therapy, and potentially earlier detection of respiratory complications such as developing atelectasis and pneumonia since ICU patients often get daily morning chest X-rays.

In summary, a rib fracture triage guideline that de-emphasized the patient’s age and the number of fractured ribs while considering functional status lowered ICU admission rates without compromising patient safety. Poor functional parameters were stronger predictors of complications and prolonged hospital stay than patient demographics and many anatomical considerations. A brief assessment based on functional parameters upon patient admission may play an important role in future risk stratification and triage guidelines for those with blunt chest wall trauma.

Authors’ Note

Presented at annual meeting of the Southeastern Surgical Congress in New Orleans, LA in February 2020.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.