Patient Reported Outcomes Following Severe Abdominal Trauma: A Secondary Analysis of the DCL Trial

Abstract

Objective:

Little is known about patient-reported outcomes (PROs) following abdominal trauma. We hypothesized that patients undergoing definitive laparotomy (DEF) would have better PROs compared to those treated with damage control laparotomy (DCL).

Materials and Methods:

The DCL Trial randomized DEF versus DCL in abdominal trauma. PROs were measured using the European Quality of Life-5 Dimensions-5 Levels (EQ-5D) questionnaire at discharge and 6 months post-discharge (1 = perfect health, 0 = death, and < 0 = worse than death) and Posttraumatic Stress Disorder (PTSD) Checklist-Civilian. Unadjusted Bayesian analysis with a neutral prior was used to assess the posterior probability of achieving minimal clinically important difference (MCID).

Results:

Of 39 randomized patients (21 DEF vs 18 DCL), 8 patients died (7 DEF vs 1 DCL). Of those who survived, 28 completed the EQ-5D at discharge (12 DEF vs 16 DCL) and 25 at 6 months (12 DEF vs 13 DCL). Most patients were male (79%) with a median age of 30 (IQR 21–42), suffered blunt injury (56%), and were severely injured (median injury severity score 33, IQR 21 – 42). Median EQ-5D value at discharge was 0.20 (IQR 0.06 – 0.52) DEF vs. 0.31 (IQR −0.03 - 0.43) DCL, and at 6 months 0.51 (IQR 0.30 – 0.74) DEF vs. 0.50 (IQR 0.28 – 0.84) DCL. The posterior probability of MCID DEF vs DCL at discharge and 6 months, was 16% and 23%, respectively.

Conclusions:

Functional deficits for trauma patients persist beyond the acute setting regardless of laparotomy status. These deficits warrant longitudinal studies to better inform patients on recovery expectations.

Introduction

In trauma surgery, damage control laparotomy (DCL) was introduced as a temporizing technique to salvage severely injured patients and to mitigate the sequalae of the “trauma triad” – coagulopathy, hypothermia, and acidosis.^1–6 This is in contrast to definitive laparotomy (DEF) where all injuries are repaired and the abdomen is closed at the index surgery.⁷ DCL utilizes three stages to provide definitive care to patients – 1) emergency laparotomy to control contamination and hemorrhage with temporary abdominal closure, 2) transfer to the intensive care unit for correction of the trauma triad, and 3) reoperation for definitive repair of injuries and abdominal closure.^1,8,9 Published comparative studies of DCL in trauma are limited both in quantity and quality secondary to inherent limitations from observational studies.^7,10,11 Furthermore, measured outcomes largely focus on mortality and morbidity (failure of abdominal wall closure, ventral hernias, enterocutaneous fistula, etc.) while little is known about the long term sequalae of disability and diminished quality of life following abdominal injury.^12–17

Data on patient reported outcomes (PROs) are rare.^1,18 PROs improve surgical care by representing the perception and views of patients without interpretation by investigators, aiding surgeons in providing patient-centered care.^12,19–24 While the number of PROs published in trauma has increased over the last 3 decades, it has not kept pace with PRO studies in general populations.¹² Furthermore, it is unclear which instrument is optimal to measure trauma-related PROs.^12,25

Our index study was a randomized controlled trial comparing DEF versus DCL among trauma patients with relative indications for damage control.²⁶ The goals of this analysis were to evaluate post-discharge PROs in a severely injured cohort of trauma laparotomy patients and to identify differences in PROs after DEF versus DCL. We hypothesized that patients undergoing DEF would have better PROs than patients undergoing DCL.

Methods

The research protocol for the DCL Trial was approved by the Institutional Review Board (HSC-GEN-16–0104) and registered with clinicaltrials.gov (identifier NCT02706041).²⁷ The protocol and the outcomes of the DCL Trial have been previously published.^26,27 Briefly, this was a single-center randomized controlled trial to compare the effect of DEF versus DCL in trauma patients ≥ 16 undergoing emergent laparotomy. Indications for DCL for enrollment in the trial included: planned second laparotomy for evaluation of ischemia or contamination, need to expedite transport to computed tomography or to the intensive care unit, and isolated metabolic acidosis in the absence of ongoing transfusions or hypotension. Patients for whom there was an absolute indication for DCL such as need for gauze packing for liver or retroperitoneum hemorrhage, need for interventional radiology for hemorrhage control, abdominal compartment syndrome prophylaxis, hemodynamic instability, or those wearing an opt-out bracelet, were excluded. Other specific inclusion and exclusion criteria have been described in detail in the published protocol.²⁶ Enrolled patients were randomized to either DEF or DCL. The trial enrolled patients from June 2016 to May 2019.

PROs were measured using the previously validated European Quality of Life-5 Dimensions-5 Levels (EQ-5D) questionnaire and the Posttraumatic Stress Disorder (PTSD) Checklist-Civilian (PCL-C) which were administered at discharge and 6 months post-discharge.^28,29 The 6-month post-discharge interview was conducted via telephone. Although not specific for trauma, the EQ-5D questionnaire is popular tool to study PROs across different surgical subspecialties.

The EQ-5D survey uses individual patient responses to each question (levels); these responses were compiled into states and converted to value scores according to the United States valuation scale.³⁰ The five dimensions in the EQ-5D survey are mobility, self-care, usual activities, pain and discomfort, and anxiety and depression. The five levels used in the EQ-5D survey are no problems, slight problems, moderate problems, severe problems, and extreme problems. A value score of 1 indicates perfect health, a value score of 0 represents death, and a value score < 0 represents an outcome perceived to be worse than death. Patients who died prior to discharge were assigned a value of zero. The estimated minimally clinically important difference (MCID) was set as 0.05.³¹ Unadjusted Bayesian analysis with a neutral prior was performed to assess the posterior probability of achieving the MCID.

It is well established that patients requiring intensive care, especially as a result of a traumatic event, frequently experience symptoms of (PTSD).^32–34 It is unclear if leaving an abdomen open, as in DCL, would add to the traumatic experience. To assess the differences in PTSD severity between groups, the PCL-C, a measure to assess PTSD severity, was administered at hospital discharge and at 6 months.^28,35 A PCL-C score of 17–27 indicates little/no severity, 28–29 some PTSD symptoms, 30–44 moderate to moderately high severity PTSD, and a score of 45–85 a high severity of PTSD.

An intention-to-treat analysis was performed. Due to small sample size and limitations of frequentist analyses, an unadjusted Bayesian analysis with a neutral prior was performed to assess the posterior probability of achieving the MCID between groups for EQ-5D.^36–39 All analyses were performed in R version 4.1.2 (R Core Team. 2021. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria)

Results

Of the 39 patients who consented to outcome data analysis, 21 underwent DEF and 18 underwent DCL. There were no clinically significant differences in patient demographics or specific injuries between groups. While both groups suffered severe trauma (median Injury Severity Score for entire group 33 (IQR 20–43), the DEF group had a higher median Injury Severity Score. (Table 1) Eight patients died (7 DEF vs 1 DCL) prior to discharge and were assigned an EQ-5D value of 0. Of those who survived, 28 patients completed the EQ-5D questionnaire at discharge (12 DEF vs 16 DCL) and 25 patients at 6 months (12 DEF vs 13 DCL). (Figure 1) There was no significant difference in major abdominal complications or death between the two groups.

Table 1:

Demographics and injury severity. Continuous data presented as median (interquartile range). Categorical data presented as number (%). DCL = damage control laparotomy. DEF = definitive laparotomy. AIS = Abbreviated Injury Scale

	DEF (n = 21)	DCL (n = 18)
Demographics
Age, years	29 (24–48)	32 (28–40)
Sex
Female	4 (19)	4 (22)
Male	17 (81)	14 (78)
Mechanism
Blunt	12 (57)	10 (56)
Penetrating	9 (43)	8 (44)
Injuries
Liver	10 (48)	10 (56)
Spleen	11 (52)	10 (56)
Kidney	7 (33)	6 (33)
Small Bowel	9 (43)	8 (44)
Large Bowel	7 (33)	6 (33)
Pancreas	7 (33)	3 (17)
Stomach	5 (24)	3 (17)
Pelvis	8 (38)	7 (39)
Traumatic Brain Injury	6 (29)	5 (28)
Injury Severity
Head AIS	1 (0–3)	0 (0–4)
Chest AIS	3 (0–3)	3 (1–3)
Abdomen AIS	3 (3–4)	4 (3–4)
Injury Severity Score	34 (20–43)	29 (22–41)

Figure 1:

Flow chart of survey completion by treatment arm. DEF = definitive laparotomy, DCL = damage control laparotomy.

At discharge, the median EQ-5D value was 0.20 (IQR 0.06 – 0.52) and 0.31 (IQR −0.03 −0.43) for DEF and DCL, respectively. At 6 months, median EQ-5D values were 0.51 (IQR 0.30 – 0.74) and 0.50 (IQR 0.28 – 0.84) for the DEF and DCL, respectively. There was no significant difference between the groups at discharge or 6 months post-discharge. (Figure 2) The posterior probability of a MCID in EQ-5D in DEF vs DCL at discharge and 6 months, was 16% and 23%, respectively. Although there was improvement in EQ-5D value between discharge and 6 months post-discharge, both groups failed to report return to baseline status at 6 months post-discharge.

Figure 2:

Differences between EQ-5D scores of treatment groups. DCL = damage control laparotomy, DEF = definitive laparotomy

When assessing EQ-5D dimensions individually, more than half of patients reported moderate to extreme problems in most dimensions at discharge. At 6 months, patients continued to report persistent deficits, particularly usual activities and pain and discomfort (Figure 3).

Figure 3:

Percentage of patients reporting levels within EQ-5D dimensions. DEF = definitive laparotomy, DCL = damage control laparotomy.

The PCL-C was completed by 27 patients (3 DEF vs 9 DCL) at discharge and 25 patients (6 DEF vs 9 DCL). There was no difference in score at discharge (DEF 25 (IQR 21–33) vs DCL 32 (IQR 23–44)) or at 6 months (DEF 32 (25–40) vs DCL 29 (23–45)). Symptoms of PTSD persisted at 6 months in both groups.

Discussion

The results of our trial highlight that PROs remain persistently poor at 6 months following discharge in severely injured trauma patients undergoing emergent laparotomy, regardless of DEF or DCL. In this limited cohort of patients, we were unable to identify a difference in PROs in patients managed with DEF versus DCL. At 6 months, median EQ-5D value score for DEF and DCL were 0.51 and 0.50, respectively, indicating that patients had yet to return to perfect health. Encouragingly, when compared to EQ-5D value score at discharge, both treatment groups demonstrated improvement. Similar to the EQ-5D, we were also unable to identify a difference in PCL-C between treatment groups. However, at 6 months, persistent PTSD symptoms were reported (some symptoms in DCL, moderate-moderately high symptoms in DEF). An inverse relationship between presence of PTSD symptoms and quality of life has also been described in the literature.^34,40–46 These results draw attention to the fact that longitudinal PRO studies in trauma are warranted to better understand long term implications and to better provide resources to assist patients.

Traditional surgical outcomes studies have focused on “audit measures”, such as mortality, morbidity, duration of hospital length of stay, or other objective measures.^47,48 These measures are directly observable and quantifiable. Clinically, surgeons may find interest in these outcomes, however patients equate recovery as absence of symptoms and returning to activities of daily living.⁴⁷ These surgical outcomes are not a reliably proxy for patient recovery because they may only persist for a short term, whereas non-surgical outcomes may persist for longer periods of time.⁴⁷ Our results show that approximately 40% of patients report severe to extreme problems within usual activities and at least 20% of patients continue to have at least moderate problems with all categories. This highlights the longitudinal deficits that patients face when they are long beyond the acute care phase.

Admittedly, PROs are difficult to measure.⁴⁹ One reason they are difficult to measure is secondary to a lack of high-quality previously-validated PRO instruments. In a systematic review of measurement properties of PROs after abdominal surgery, Fiore et al found 22 PRO measures with a majority of measures having deficiencies.⁵⁰ The authors could not recommend a specific PRO instrument for abdominal surgery. Within the scope of trauma, PRO studies are largely dominated by orthopedic trauma, traumatic brain/spinal cord injury, and burns.¹² EQ-5D, one of the PRO instruments used in our study, was one of the most common PRO measures utilized.¹² EQ-5D is validated measure of general health, however it lacks measures specific for trauma. Trauma-specific measures are important as they accurately demonstrate changes to trauma-specific interventions.^12,25 However, trauma-specific PRO instruments have been previously described, yet use in published studies has been limited and plagued with considerable variability.^12,25,51,52

PROs should be collected and integrated into patient discussions to better describe health trajectory, particularly following traumatic injury.^12,53,54 It has been reported that trauma patients suffer cognitive, somatic, or psychological deficits as far as 1–2 years post-injury.^45,55–62 Additionally, risk factors such as mechanism, gender, or socio-economic status have been associated with negative long-term effects on quality of life following trauma.^63,64 These deficits and factors should be considered when discussing goals of recovery with patients. The Functional Outcomes and Recovery after Trauma Emergencies (FORTE) project was established in 2015 to better understand PROs following traumatic injuries in the United States.⁵⁸ Their results show that trauma patients face persistent impairment 6 to 12 months following injury and therefore trauma registries should augment their data collection with longitudinally-collected PROs for 6 to 12 months post-injury.⁵⁸ It is important to identify duration of deficits as there may be benefit to continue post-discharge interventions such as physical therapy and counseling in this cohort of patients.

This trial had several limitations, including small size, narrow scope, and lack of blinding. This study was the culmination of 3 years of prior stakeholder-driven collaboration on identifying in which patients there was equipoise for DCL. As a result, temporal changes in management of these patients may have played a role in the selection of patients. Importantly, of patients who died, the majority (75%) elected to transition to comfort care. It is difficult to quantify these decisions as they play a role in PROs. Although we gave these patients an EQ-5D value score of 0, in reality they chose that their current status is worse than death. Future studies in PROs will need to identify how to best address this issue.

Conclusion

Our results demonstrate that PRO studies are feasible in trauma patients. Furthermore, our results highlight the persistent functional and emotional deficits trauma patients face as far as 6 months following injury, regardless of DEF or DCL. Since these functional deficits persist longer than previously anticipated, there may be benefit in extending post-discharge care and interventions longer than previously anticipated.