Introduction

Emergency laparotomy is a time-sensitive procedure with a high mortality range from 8.8 to 18.6% [1,2,3], which is much higher than elective surgeries [4,5,6]. Elderly patients suffering worse morbidity and mortality following emergency laparotomy have been contributed to multiple medical comorbidities and nutrition status [7]. Effective identification of high-risk patients has been the key to improving perioperative outcomes [5]. However, due to the complexity of the patient's condition, the different variety of surgical types, and the tight operation preparation time, it has been challenging to predict the outcome after emergency surgery based on preoperative information. Several previous studies have used various scoring systems to predict the risk of emergency laparotomy [8, 9]. However, when patients’ condition is too critical to complete functional tests and answer related questions in emergency conditions, the scores are usually subjective and inaccurate [10, 11]. Therefore, it is an urgent requirement to develop a new assessment tool to identify the patient at risk of emergency laparotomy and guide optimal perioperative management [9]. Body composition also plays an important role in predicting treatment outcomes in patients following surgery. Sarcopenia has been observed to be a strong prognostic indicator for perioperative complications [12, 13], including cognitive impairment [14], fractures [15], mental disorders [16], and even survival [17, 18].

Sarcopenia refers to the progressive and global decline in skeletal muscle mass and strength associated with aging, immobility, or illness status [19]. Although there are discrepancies in the diagnostic criteria of sarcopenia in different countries and regions, the cross-sectional area of the lumbar muscle on an abdominal computed tomography (CT) scan is an internationally recognized simple and reliable indicator [20]. It is assessed by measuring muscle mass at the level of the L3 vertebra, which has made a preoperative assessment of psoas major area (PMA) and total skeletal muscle area (SMA) possible based on the routine examination of an abdominal CT scan [21].

Studies have shown that sarcopenia increases the incidence of adverse events and mortality after elective esophageal cancer surgery, gastrectomy, and pancreatic surgery [22,23,24]. However, there has been no definite conclusion on the impact of 30-day mortality and postoperative adverse events on emergency laparotomy. Hajibandeh et al. pointed out that sarcopenia can be used to predict mortality in both emergency and elective abdominal surgeries [25]. However, only four studies of emergency surgery were included, so it is unconvincing to draw reliable conclusions. We, therefore, performed further analysis to assess the impact of sarcopenia on 30-day mortality and postoperative complications in patients following emergency laparotomy.

Methods

This systematic review and meta-analysis were prepared in accordance with the latest PRISMA requirements and was registered with Prospero (registration number: CRD42022300132) [26]. Two researchers (T.Y. and K.L.) searched databases such as PubMed, Embase, and Cochrane. The search date was as of March 1, 2022. The search was not limited to language and region, and we provided a PRISMA checklist. PubMed's search strategy can be found in Table 1.

Table 1 Search strategy of PubMed
Full size table

Study selection

We aimed to include all the studies comparing patients with sarcopenia and non-sarcopenic patients following emergency laparotomy. Inclusion and exclusion criteria were conducted in advance. The inclusion criteria were as follows:

  1. 1.

    Age ≥ 18 years old;

  2. 2.

    Patients were treated by emergency laparotomies;

  3. 3.

    Preoperative abdominal/pelvic CT data were present.

Emergency operations in this study included segmental or total colectomy, small bowel resection, open cholecystectomy, open appendectomy, abscess drainage, exploratory laparotomy, etc. Our exclusion criteria were:

  1. 1.

    Elective surgery;

  2. 2.

    Traumatic abdominal surgery;

  3. 3.

    Emergency abdominal vascular surgery;

  4. 4.

    Studies with an unclear diagnosis of sarcopenia.

Two of our investigators (T.Y. and K.L.) selected studies that were compliant with a full-text reading by reviewing titles and abstracts. Any disagreements between investigators were independently resolved (R.W.). In addition, we searched the World Health Organization International Clinical Trials Registry and queried bibliographic lists of relevant articles and reviews for further potentially eligible studies.

Data extraction

Two investigators (T.Y. and X.D.) independently extracted the following data: author name and publication year, literature type, the sample size for exposure and control groups, diagnostic criteria for sarcopenia, and characteristics of the included population. During this process, all disagreements were resolved by discussion, and if necessary, a third author (P.J.) was consulted.

We extracted data directly from the original text for synthesis. If the data were presented in the form of a graph and could not be directly extracted, we used a Plot digitizer or contacted the corresponding author. We extracted continuous results as the mean and standard deviation, and if the median was displayed, we converted the median and interquartile range to mean and standard deviation using the statistical formula [27, 28].

Quality assessment and risk of bias

Two investigators (T.Y. and K.L.) independently used the Quality in Prognosis Studies (QUIPS) critical assessment tool to assess the risk of bias for including studies [29, 30]. This tool is designed for systematic reviews of prognostic factor studies. The scale mainly includes study participation, study attrition, prognostic factor measurement, outcome measurement, study confounding, and statistical analysis and reporting. Each domain is assessed against criteria, thereby resulting in a rating of ‘high,’ ‘moderate,’ or ‘low’ risk of bias. Any discrepancies between investigators were discussed at a consensus meeting, and any further disagreement was resolved by discussion with a third investigator.

We used the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) [31] approach to assess the quality of evidence for 30-day postoperative mortality, complication rates, ICU admission, the length of ICU stay, and the length of hospital stay. We rated the quality of the evidence as 'high', 'moderate', 'low', and 'very low' based on risk of bias, inconsistency, indirectness, imprecision, and other considerations. And we used GRADEpro to generate the Summary of Finding (SoF).

Outcome

The primary outcome was 30-day mortality after emergency laparotomy. Secondary outcomes included incidence of postoperative complications, number of postoperative ICU admissions, and ICU and hospital length of stay, respectively.

Data analysis

We used Revman 5.3 for meta-analysis. For continuous variables, we used mean difference (MD) and 95% confidence interval (CI). For binary variables, the odds ratio (OR) value for statistics was implemented. In this study, considering that there was always heterogeneity in terms of the type of surgery, surgical technique, and experience of surgeons, all results in this study were performed using a random-effects model. For assessing the outcome, we performed a sensitivity analysis by using the leave-one-out approach to identify the possible sources of heterogeneity.

Result

Through systematic database searching, we identified 300 articles. After screening to remove duplicate literature, there were 260 articles in total. Thirty-three articles were reviewed for full text following evaluation of the titles and abstracts. Twenty-two articles were excluded due to non-emergency surgery and unclear diagnosis. Review and conference abstract were also excluded in terms of the study design. Overall, 11 articles with 3795 patients were included for further analysis. Fig 1 depicts a flowchart of the study selection process.

Fig 1.
figure 1

Flowchart showing selection of articles for review

Full size image

Table 2 summarizes the characteristics of the 11 articles that met the inclusion criteria. The total sample size ranged from 80 to 967, all of which were retrospective studies [5, 17, 18, 32,33,34,35,36,37,38,39]. Of the studies, 10 of them examined muscle mass at the level of the L3 vertebral body on CT scans to diagnose sarcopenia [5, 17, 32,33,34,35,36,37, 39] and only one assessed the muscle mass at the level of the L4 vertebral body on CT scans [38].

Table 2 Characteristics of included studies
Full size table

Risk of bias

Table 3 shows the risk of bias for the 11 included studies assessed according to the QUIPS tool, of which almost all were rated moderate to high for potential risk of bias in study attrition and study confounding domains. This was the most common methodological weakness.

Table 3 Risk of bias summary: judgment of each domain for all included studies using the Quality of Prognostic Studies (QUIPS) tool
Full size table

Quality of evidence

Based on the GRADE approach, we found the moderate quality of evidence for 30-day postoperative mortality and length of hospital stay. The quality of evidence was low for the need for ICU admission and the incidence of postoperative complications, and the quality of evidence for the length of stay in the ICU was very low (Fig. 2).

Fig 2.
figure 2

Certainty of the evidence and summary of findings

Full size image

Primary outcome

30-day mortality

Nine articles with a total of 3626 patients reported 30-day mortality following emergency laparotomy [5, 17, 32,33,34, 36,37,38,39]. Compared with non-sarcopenic patients, sarcopenic patients had a higher risk of death 30 days after surgery (Fig. 3; OR = 2.42, 95% CI = 1.93–3.05, P < 0.00001). There was slight heterogeneity among the included kinds of literature (I2 = 8%, P = 0.37).

Fig 3.
figure 3

Forest plot showing 30-day mortality

Full size image

Secondary outcome

Length of ICU stay

There were four articles documenting the length of stay in the ICU, in which data were reported for 1374 cases [5, 34,35,36]. The sarcopenia group was found to have longer ICU stays (Fig. 4; MD = 0.55, 95% CI = 0.05–1.06, P = 0.03), with acceptable heterogeneity between articles (I2 = 47%, P = 0.13).

Fig 4.
figure 4

Forest plot showing the length of ICU stay

Full size image

Need for ICU admission

Four articles with a total of 1385 patients documented the number of postoperative ICU admissions required [5, 34, 36, 37]. There was a significant difference between sarcopenia and non-sarcopenic groups (Fig. 5; OR = 1.58, 95% CI = 1.11–2.25, P = 0.01). Patients diagnosed with sarcopenia were more likely to be admitted to the ICU after emergency surgery. Among all included articles, the identity was high and there was no heterogeneity (I2 = 0%, P = 0.54).

Fig 5.
figure 5

Forest plot showing the need for ICU admission

Full size image

Total complications

A total of six articles reported the incidence of postoperative complications in the emergency department, with a total of 2737 patients [5, 17, 35,36,37,38]. There was low heterogeneity among the included articles (I2 = 33%, P = 0.19). Patients with sarcopenia were more likely to have certain complications after emergency laparotomy (Fig. 6; OR = 1.78, 95% CI = 1.41–2.26, P < 0.00001).

Fig 6.
figure 6

Forest plot showing the incidence of total complications

Full size image

Length of hospital stay

There were nine articles with a total of 3565 patients reporting postoperative hospital stays [5, 17, 32,33,34,35,36,37,38]. There was a significant difference in the length of hospital stay between the sarcopenia group and the non-sarcopenic group (Fig. 7; MD = 2.33, 95% CI = 1.33–3.32, P < 0.00001; I2 = 56%). This suggested that patients with sarcopenia had longer hospital stays after emergency surgery. Due to significant heterogeneity, we further performed a sensitivity analysis. In the study by Samer, only the duration of hospital stay in patients who survived within the first 30 days after surgery was counted [32]. After excluding this article, there was still a significant difference in the length of hospital stay between the two groups, but the heterogeneity between articles was reduced (Table 4; MD = 1.94, 95% CI = 1.23–2.65, P < 0.00001; I2 = 30%).

Fig 7.
figure 7

Forest plot showing the length of hospital stay

Full size image
Table 4 The sensitivity analysis of the length of hospital stay
Full size table

Discussion

Emergency laparotomy surgery had high morbidity and mortality. There were few studies investigating the impact of sarcopenia in patients following emergency laparotomy. Our study aimed to assess the risk of death after emergency laparotomy in patients with preoperative sarcopenia. We performed a systematic review and meta-analysis. Eleven eligible studies involving 3795 patients were included. The results showed that sarcopenia was associated with increased postoperative 30-day mortality. There was mild heterogeneity among the nine studies, which were considered high quality. In addition, patients with sarcopenia significantly increased postoperative ICU duration, number of ICU admissions, incidence of postoperative complications.

Patients with preoperative sarcopenia had significantly longer hospital length of stay, but there was large heterogeneity between studies. One of the studies included patients who survived more than or equal to 30 days after surgery in the overall length of stay calculation. The author believed that the impact of early death should be excluded. However, patients that died during the study period were included so as to not overestimate the total duration result. In fact, hospital length of stay in this study was significantly longer than others, which showed that overstated time was a variable of heterogeneity in our study. Considering ICU admissions and hospital length of stay, it was reasonable to presume that patients with sarcopenia following emergency laparotomy were at risk of high medical costs and excessive heavy illness burden.

Our systematic review and meta-analysis independently investigated the association of sarcopenia with prognosis after emergency laparotomy. An efficient and simple assessment tool to identify high-risk surgical patients may be extremely valuable for medical teams’ awareness, especially in time-sensitive situations. Our findings were consistent with previous observative studies confirming that sarcopenia was associated with increased 30-day postoperative mortality in patients undergoing emergency laparotomy [25, 40, 41]. Some studies identified the effect of sarcopenia on mortality and morbidity after elective and emergency abdominal surgeries. However, the quality of evidence is low due to the small number of included studies and participants [25]. In recent years, the impact of sarcopenia on postoperative outcomes has received extensive attention, especially in emergency laparotomy. Therefore, we performed an update of this topic including 11 studies of 3795 patients, and we used the QUIPS and GRADE tools, respectively, to assess the risk of bias and quality of evidence in the included studies. In our study, heterogeneity was low with a dramatically expanded sample size, and the methodological quality of the studies in our review was reliable, increasing the representativeness, and generalizability of our conclusions.

Previous studies found that sarcopenia was associated with multiple adverse outcomes, including falls, functional decline, and postoperative mortality [42]. Sarcopenia was assessed by abdominal computed tomography (CT) L3 pyramid or L4 pyramid total psoas area (TPA) or total psoas index (TPI). TPA < 3.64 cm/m2 in women and TPA < 4.55 cm/m2 in men or TPI < 1.50 cm/m2 in women and TPI < 2.16 cm/m2 in men were defined as sarcopenia [43,44,45,46]. Therefore, sarcopenia by abdomen CT scan was used to easily identify the high-risk patients without evaluating them from complicated scales or questionnaires, which is both costly for time and inaccurate in terms of patients’ severeness and degree of cooperation. Meanwhile, most of the patients treated by emergency laparotomy had abdominal CT scan before surgeries. Therefore, sarcopenia was simple, objective, and efficient to be a perioperative risk stratification tool in emergency clinical practice.

The reasons for poor prognosis of patients with sarcopenia after emergency laparotomy were multifactorial, including preoperative frailty, previous malnutrition, and complex comorbidities. In addition, emergency laparotomy surgeries were often insufficiently prepared due to rapid and even life-threatening disease progression. In the postoperative stage, surgical strikes, pain, respiratory failure, circulatory compromise, sepsis, and multiple organ dysfunction also increased the incidence of postoperative complications and mortality.

Although CT-identified sarcopenia can help to predict perioperative risk, we cannot improve patient outcomes through preoperative nutrition and physical activity, as practiced in elective surgery [47]. To these patient groups, we likely pay more attention throughout treatment, such as complications prevention, nutrition intervention in early post-procedure phase, and even immediate intensive care post-surgery. Therefore, preoperative diagnosis of sarcopenia undoubtedly provided important predictive information for general medical management strategies, nursing goals, family awareness, and rehabilitation expectations [48].

There were some limitations in our study. Firstly, the 11 included studies were all retrospective cohort studies. The retrospective design caused them to solely assess muscle mass without measurement of muscle function (grip strength, walking speed, physical activity), which may exaggerate the predictive power of sarcopenia. In our opinion, given that sarcopenia is a serious disease state, we would rather overestimate than underdiagnose. Secondly, there were no clear definition and classification of surgical types in the included literature, so we could not perform subgroup analysis according to surgical procedures specifically, which may have influenced the results of this study. Further research is required to be addressed on specific procedures later. In addition, positive results were more likely to be published and there may be a risk of reporting bias. Most important, in the emergency setting, we have demonstrated that sarcopenia can effectively predict adverse postoperative outcomes.

Recently, some studies argued that patients with sarcopenia cannot be diagnosed by preoperative CT efficiently because of the costs, radiations, and body position restrictions. The point-of-care ultrasound had the advantages of high operability, repeatability, and convenient portability (49). Therefore, point-of-care ultrasound instead of CT scan may be a future research direction to assess the feasibility of sarcopenia diagnosis before emergency surgeries. Exploration of randomized controlled trials is also needed if early targeted interventions on sarcopenia can improve patient outcomes based on our prediction findings.

Conclusion

We found that preoperative CT scan-derived sarcopenia was associated with increased postoperative 30-day mortality and ICU admissions. ICU and hospital length of stay and incidence of postoperative adverse events were also significantly elevated. Further research should demonstrate if dedicated intervention for sarcopenia will improve patient outcomes.