Emergency laparotomy (EL) is one of the most frequently performed yet highest-risk surgical procedures worldwide. Compared to elective abdominal surgery, EL is associated with a significantly higher complication rate—exceeding 50%—and a mortality rate up to five times greater (1, 2). Efforts by the Enhanced Recovery After Surgery (ERAS) Society (3, 4) have marked important progress toward evidence-based emergency perioperative care. Yet, the scarcity of high-quality, EL-specific evidence continues to impede the development of robust guidelines.

Unlike elective patients, who typically undergo surgery in an optimized physiological state, EL patients face the compounded physiological stress resulting from both the acute underlying pathology and the added burden of surgical trauma (5). This dual insult disrupts neuroendocrine, metabolic, and immune homeostasis, heightening susceptibility to complications that cannot be reliably predicted using elective surgery data (6). Numerous studies have identified the emergent nature of surgery itself as an independent risk factor for adverse outcomes, including postoperative pulmonary complications (PPCs), which are among the most common complications following EL (1, 7, 8).

In elective abdominal surgery, PPC incidence ranges from <1 to 9%, depending on definitions and cohort characteristics (9–11). In stark contrast, the multicenter prospective ALPINE study reported a 48% PPC incidence in EL patients (12). These complications—most commonly respiratory failure, atelectasis, and infection—substantially impair recovery and are associated with increased intensive care unit (ICU) admissions, reintubation rates, prolonged mechanical ventilation, and higher mortality (8, 9).

Opportunities for PPC prevention in EL are limited, as preoperative strategies commonly used in elective settings—such as respiratory physiotherapy or structured prehabilitation programs—are largely unfeasible in the acute context (13, 14). Consequently, intraoperative and early postoperative management—including ventilation strategies, pulmonary care bundles, and timely antibiotic therapy when a respiratory infection is suspected —play a central role. Yet, evidence from elective contexts cannot be applied uncritically: for instance, lung-protective ventilation failed to reduce PPC incidence in EL patients in the ALPINE study (12, 15).

Antibiotic management of PPCs in EL remains understudied (16). In a cohort of 7,275 abdominal surgery patients, nearly half of those who developed postoperative pneumonia received inappropriate antibiotic treatment—significantly increasing pneumonia-specific mortality (17). The challenge is exacerbated by ongoing uncertainty around optimal respiratory sampling methods in critically ill patients and the rising prevalence of antimicrobial resistance (18). Empiric antibiotic selection requires both accurate patient-specific risk stratification and familiarity with local resistance patterns (19).

Notably, preoperative colonization with multidrug-resistant organisms (MDROs) has been linked to higher complication rates, including PPCs, across various surgical cohorts (20–22). However, its specific role in EL remains unclear.

Addressing these knowledge gaps is essential. A nuanced understanding of how acute pathophysiological and surgical stressors contribute to PPC risk may enable the development of more targeted, EL-specific management strategies. In particular, identifying modifiable risk factors and elucidating the impact of MDRO colonization could inform preventive measures, optimize the timing of interventions, and support improved antibiotic stewardship in EL patients.

Accordingly, this study had two primary objectives: to identify potentially modifiable pre- and intraoperative predictors of PPCs in the setting of EL, and to characterize the spectrum of MDRO colonization most likely to impair postoperative pulmonary recovery.

This single-center, observational study retrospectively analyzed data from 928 patients who underwent EL between January 2014 and December 2020 at the Department of General, Visceral, Thoracic, and Transplant Surgery, University Hospital of Giessen. All data were drawn from prospectively maintained institutional databases.

The study was conducted in accordance with the Declaration of Helsinki and approved by the local ethics committee of the Medical Faculty at the University of Giessen (Approval No. 20/22). Given the retrospective design, the requirement for written informed consent was waived. Data collection, manuscript preparation, and submission adhered to the Committee on Publication Ethics (COPE) and the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (23).

Our study highlights the substantial clinical burden of PPCs in patients undergoing EL, emphasizing their strong association with both short-term morbidity and 90-day mortality. Through the identification of modifiable and non-modifiable preoperative risk factors—specifically preoperative hemoglobin <10 g/dL and colonization with MDROs—our findings suggest opportunities for targeted interventions even within the narrow therapeutic window inherent to emergency settings.

The PPC rate in our cohort was 35.3%, aligning with the 20–40% range reported in earlier studies and slightly lower than the 48% observed in the ALPINE trial (2, 12). Prior work by Havens et al. (1) demonstrated that among all postoperative complications—including pulmonary, cardiac, septic, and wound-related—pulmonary complications were most strongly associated with the emergent nature of surgery, reinforcing the idea that both acute surgical pathology and the procedural trauma significantly heighten respiratory risk in this particularly vulnerable patient group.

To identify actionable factors linked to both acute surgical pathology and the procedural EL trauma, we employed PSM to control for non-modifiable demographic and preexisting health issues. Our analysis revealed ASA score, CAR, preoperative hemoglobin levels, and colonization with MDROs as independent predictors of PPC development. Importantly, all of these factors were also significantly associated with 90-day mortality among patients who developed PPCs, underscoring their clinical relevance beyond the immediate postoperative period.

The ASA classification remains the most widely used and accessible tool for perioperative risk stratification (35). Its simplicity makes it preferable to more complex scoring systems, and numerous studies—predominantly in elective surgery—have demonstrated its ability to predict postoperative morbidity, including the risk of PPCs (36, 37). However, its utility in emergency contexts is debated. As Horvath et al. (38) have pointed out, the ASA system lacks clear guidance on how acute physiological insults should be integrated into the preoperative status assignment. For instance, a patient with obesity and well-controlled diabetes would typically be assigned ASA II, but if presenting with sepsis, the patient may be reclassified as ASA IV-E. While such transitions are relatively straightforward, more nuanced clinical deterioration can introduce subjectivity into the scoring process, raising concerns about its consistency and predictive power in emergency settings.

Nonetheless, in our cohort, the ASA score remained one of the most powerful independent predictors of PPCs. Patients with ASA scores above 3 experienced significantly higher 90-day mortality, both in those who developed PPCs (p = 0.0003) and in those who did not (p = 0.0012)—suggesting that despite its limitations, the ASA score retains prognostic value even in acute care settings.

CAR, a composite biomarker that reflects systemic inflammation through the positive acute-phase protein CRP (39) and the negative acute-phase protein albumin, which is also associated with nutritional risk (40), has shown good performance as an outcome predictor in critical illness across many studies (30–32).

In a large retrospective study involving over 11,000 surgical patients, Oh et al. (31) demonstrated that CAR measured upon ICU admission independently predicted both 30-day and one-year mortality. In oncologic surgery, where hypoalbuminemia and systemic inflammation are common, elevated CAR has been predictive of increased postoperative complications, although pulmonary complications were not specifically addressed (41, 42). Notably, in non-surgical cohorts such as COVID-19 patients, higher CAR values on admission have been linked to reduced oxygen saturation and impaired pulmonary function parameters (SO₂, FEV₁%, and FVC%) (43).

In our study, the preoperative CAR emerged as a robust independent predictor of PPCs in EL. Using the Youden index (29), we identified an optimal CAR cut-off value of 1.07, which was predictive of 90-day survival regardless of whether a pulmonary complication occurred.

Mechanistically, the prognostic value of CAR is underpinned by the pathophysiological roles of its two components—CRP, a positive acute-phase reactant, and albumin, a negative acute-phase reactant. CRP, in its native pentameric form, dissociates upon encountering damaged cell membranes into monomeric subunits that exhibit potent pro-inflammatory and pro-coagulant properties (39). As shown by Fendl et al. (44), these monomeric CRP forms are trafficked via extracellular vesicles to distant organs, thereby amplifying systemic inflammation and contributing to remote organ injury, including lung damage.

Conversely, hypoalbuminemia reduces plasma oncotic pressure, facilitating fluid transudation into the alveolar space and promoting pulmonary edema. This condition has been independently associated with an elevated risk of acute respiratory distress syndrome (ARDS) in critically ill patients (45).

Although therapeutic strategies aimed at neutralizing CRP’s pro-inflammatory activity are under investigation, they have not yet advanced beyond experimental phases (46). Similarly, the efficacy of albumin supplementation to prevent ARDS remains inconclusive (47). Therefore, while CAR itself is not a directly modifiable parameter at present, it may serve as a valuable prognostic biomarker and a potential target for future therapeutic interventions in patients undergoing EL.

Preoperative anemia with a hemoglobin level below 100 g/L has been categorized as a modifiable risk factor for PPC development in the pivotal work of Miskovic and Lumb (9). Because intraoperative blood transfusion has also been observed to increase PPC risk, the authors recommend alternative treatment strategies—such as dietary measures or iron supplementation—depending on the underlying cause of the anemia (9). Such optimization, however, is naturally limited in emergency settings.

Currently, there is limited evidence regarding the significance and clinical consequences of preoperative colonization with MDROs in patients undergoing EL. When formulating clinical practice guidelines for perioperative antibiotic prophylaxis in patients colonized with Gram-negative MDROs, the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) explicitly emphasized major gaps in the available evidence (48). As a result, the issued recommendations were primarily based on low-level data and restricted to select patient populations, such as those undergoing elective colorectal surgery (20, 48, 49), or solid organ transplantation (22, 48).

In elective colorectal surgery, preoperative colonization with MDROs has been associated with an increased risk of surgical site infections (20, 48, 49). Among solid organ transplant recipients, such colonization has been linked to more severe postoperative complications, including elevated rates of all-cause mortality, infection-related mortality, polymicrobial infections, and bloodstream infections (22, 48). In critically ill patients—regardless of surgical status—MDRO colonization at ICU admission has similarly been correlated with higher in-hospital mortality, even after adjusting for illness severity (50, 51).

These findings suggest that the clinical impact of MDRO colonization is particularly pronounced in immunocompromised or physiologically unstable patients. In such cases, delays in initiating appropriate antimicrobial therapy may substantially contribute to poor outcomes. In our study, PPC patients demonstrated a significantly higher prevalence of MDRO colonization prior to surgery (25%) compared to those without PPCs (13%). Importantly, this colonization was also associated with significantly decreased 90-day survival in the PPC group (p = 0.0359), whereas no such association was observed in patients without pulmonary complications—possibly reflecting the amplified inflammatory and infectious burden in the former.

Microbiological cultures from the upper airways were obtained in 60% of patients who developed PPCs. Interestingly, while patients with Gram-positive MDRO colonization typically developed respiratory infections with unrelated pathogens, 26.8% of those colonized with Gram-negative MDROs harbored the same organism in their postoperative airway samples. Despite this, only half of these patients had received empiric antibiotic coverage that adequately targeted these resistant pathogens. This raises an important question: could earlier, more tailored antimicrobial therapy reduce pulmonary complications and mortality in high-risk, colonized patients?

Although MDRO colonization emerged as an important preoperative predictor of PPCs in our analysis, routine microbiological screening and antimicrobial susceptibility testing are generally not available prior to emergency surgery—unless this information is already known from a previous hospitalization. Nevertheless, several rapid diagnostic platforms have shown the potential to identify MDRO colonization or infection within clinically actionable timeframes. These include MALDI-TOF/RAST (matrix-assisted laser desorption/ionization time-of-flight mass spectrometry with rapid antimicrobial susceptibility testing), multiplex culture-independent diagnostics (MCD), and point-of-care multiplex polymerase chain reaction (PCR) systems, which can provide results within approximately 0.5–7 h (52–54). Although these technologies are not yet widely available in routine emergency surgical practice, their broader future implementation may allow earlier detection of MDRO carriers and help guide more targeted perioperative antimicrobial strategies.

This study is subject to several limitations. Its retrospective design, while enabling real-world data collection, inherently restricts causal interpretation. Functional respiratory parameters and preoperative arterial blood gas analyses were not consistently available and therefore could not be included in the analysis. Microbiological airway sampling was performed only in patients with clinical indications rather than routinely, introducing a potential selection bias. As all data were collected at a tertiary university hospital, referral bias is also possible, which may limit the generalizability of our findings. Finally, regional variation in MDRO prevalence, resistance patterns, and antimicrobial stewardship practices may restrict the transferability of our results to other healthcare settings.

Despite these limitations, this study represents an important first step in systematically investigating the impact of PPCs and MDRO colonization in the high-risk EL population. By identifying both modifiable and non-modifiable risk factors associated with poor outcomes, our findings lay the groundwork for improved perioperative risk stratification and patient-centered management strategies. Given the increasing prevalence of antimicrobial resistance and the persistently high morbidity associated with PPCs in emergency surgery, further prospective, multicenter studies are urgently needed to validate these findings and inform evidence-based interventions tailored to this vulnerable patient cohort.

In summary, PPCs following EL are common and have a profound impact on patient outcomes, significantly increasing mortality. Our study identifies a high preoperative ASA score, an elevated preoperative CAR, low preoperative hemoglobin, and MDRO colonization as independent predictors of PPC development and as factors associated with increased 90-day mortality in patients with PPCs. Among these, preoperative colonization with Gram-negative MDROs may represent the most readily actionable risk factor, even in the constrained emergency setting. These findings underscore the importance of rigorous preoperative risk stratification and support future studies to evaluate whether patients at high risk of PPCs could benefit from antibiotic regimens tailored to their colonizing MDROs.