Predictors of intensive care unit admission rates in patients with acute cholangitis

Abstract

Acute cholangitis (AC) is a heterogeneous disease with considerable variation in clinical presentation and high medical costs. Although the overall mortality rate is decreasing. However, the mortality in severe AC ranged from 10%-30% due to sepsis, multi-organ failure, and systemic inflammatory response syndrome. Assessing the predictors of poor outcomes, including mortality, intensive care unit (ICU) admission, and hospital stay, is vital for early and effective intervention. We assessed the predictors of mortality and ICU admission in patients with AC. We conducted a literature search in PubMed/MEDLINE, Google Scholar, and Cochrane Library for relevant articles. The keywords used were acute cholangitis, biliary cholangitis, predictors, mortality, ICU admission, and Hospital stay. In addition, we assessed the role of early vs late endoscopic retrograde cholangiopancreatography on the outcomes. Old age, end-organ failure, red cell distribution width to albumin ratio, neutrophil-to-lymphocyte ratio, platelet-lymphocyte ratio, the need for ventilator support, diabetes, electrolyte imbalance, procalcitonin-to-albumin ratio, C-reactive protein-to-albumin, Glasgow Coma Scale, and systolic hypertension are predictors of poor outcomes in AC of varying etiology, and concurrent acute pancreatitis was not associated with ICU admission.

Key Words: Acute cholangitis; Predictors; Intensive care unit admission; Mortality; Intensive care unit admission rates

Core Tip: Acute cholangitis (AC) is the infection of the bile and bile ducts, usually resulting from obstruction by stones, benign strictures, and malignancies. AC is a heterogeneous disease with considerable variation in clinical presentation and high medical costs. Literature on AC prognosis is scarce, and studies investigating the indicators of admission to intensive care units are needed for the timely introduction of diagnostic measures and treatment. We gave an important insight regarding the independent predictors of mortality and intensive care unit admission.

TO THE EDITOR

Acute cholangitis (AC) is the infection of the bile and bile ducts, usually resulting from obstruction by stones, benign strictures, and malignancies. AC is divided into mild, moderate, and severe based on experts’ criteria. Despite the progress in its treatment, AC is serious, with mortality observed in 1%-14% of patients depending on severity[1]. AC is a heterogeneous disease with considerable variation in clinical presentation and high medical costs. Although the overall mortality rate is decreasing. However, the mortality in severe AC ranged from 10%-30% due to sepsis, multi-organ failure, and systemic inflammatory response syndrome[2,3].

AC diagnosis and severity grading are based on Tokyo Guidelines first published in 2007 and updated in the years 2013 and 2018. The guidelines are based on systemic inflammation parameters, imaging, and cholestasis with each component further divided in to two items. The definitive diagnosis is fulfilled with positive one item from all the three components, while suspected AC is considered when one item in systemic parameters and one item in either imaging and cholangitis is positive[4,5].

The treatment is antibiotics coverage and biliary drainage, with early biliary drainage significantly improving the prognosis[6]. Importantly, early endoscopic retrograde cholangiopancreatography (ERCP) is associated with mortality reduction and shorter hospital stay when performed in 24-72 hours, with no benefit when conducted in the first 24 hours[7-9].

Literature on AC prognosis is scarce, and studies investigating the indicators of admission to intensive care unit (ICU) are needed for the timely introduction of diagnostic measures and treatment[10,11]. Therefore, we conducted a literature search to assess the predictors of poor outcomes in patients with AC.

We searched PubMed/MEDLINE, Cochrane Library, and Google Scholar for articles assessing the predictors of outcomes, including mortality, hospital stay, and ICU admission in patients with AC. The terms used were AC, biliary cholangitis, predictors, mortality, ICU admission, and Hospital stay. In addition, we assessed the role of early vs late ERCP on the outcomes.

PREDICTORS OF MORTALITY AND ICU ADMISSION IN PATIENTS WITH AC

The predictors of mortality and ICU admission are multifactorial and are affected by the type of hospital, patient’s characteristics including end-organ damage, complete blood count parameters and ratios, patient’s immunity, and sepsis biomarkers. In addition, the etiology of AC, and type and time of intervention significantly affect the outcomes.

Li et al[12] investigated the immune-nutritional markers and inflammatory markers in patients with AC and found that neutrophil-to-lymphocyte ratio (NLR) and prognostic nutritional index are better indicators of AC severity, supporting the findings of Yesil et al[13], who conducted a retrospective study including 633 patients with AC and found an association between platelet-lymphocyte ratio (PLR), NLR, and AC severity. A study conducted in Japan[14] included 699 patients with liver cirrhosis, who were observed to have liver cirrhosis, which is a poor indicator of AC and is associated with high bilirubin and poor albumin recovery. Kozai et al[15] conducted a recent study and found that serum procalcitonin might be a pointer of severity in malignancy-associated AC, but there was no predictive value of procalcitonin in other types of AC. Yagnik et al[16] conducted a study and found that patients with AC admitted to teaching hospitals have higher mortality compared to those admitted to non-teaching hospitals. The authors explained their findings by the higher resident and fellow autonomy in teaching hospitals and the fact that teaching hospitals admitted complex, higher-acuity AC cases.

Huang et al[17] introduced a machine-learning-derived online prediction models for the outcomes with high sensitivity and specificity in which meet the precision requirements of modern diagnosis and treatment compared to the previous models by Pan et al[7], who found that the Glasgow Coma Scale (GCS), oxygen saturation, aspartate transaminase/alanine transaminase ration, serum albumin, glucose, potassium, prothrombin time, and peripheral vascular disease are independent risk factors of severity in AC. Liu et al[18] introduced age, diabetes, ventilator-support time, systolic hypertension, and coagulopathy as indicators for sepsis and mortality, while Inan et al[19] found that old age, low albumin, low platelets count, and low lymphocytes are predictors of severity. Sheng et al[20] in their parsimonious model used sodium, albumin, red cell distribution width, hemoglobin, and the Charlson comorbidity index to predict inflammation, multi-organ failure, and severity. The authors divided AC into inflammatory and non-inflammatory dysfunction types and showed that the inflammatory dysfunction subtype is associated with poor outcomes. The poor outcomes of malignant biliary obstruction compared to biliary stone are obvious, and malignant obstruction is associated with abnormal liver and renal function tests, abnormal white blood cells, and low platelets[21]. Acehan et al[22] found that malignant etiology, creatinine, and red cell distribution width to albumin ratio (RAR) were independent predictors of ICU admission and mortality, RAR score of 3.8 achieved, and sensitivity and specificity of 94.1% and 56.7% respectively, and higher than Tokyo Severity Grading and other prognostic indicators, procalcitonin-to-albumin ratio (PAR), and C-reactive protein to procalcitonin ration for predicting ICU admission.

THE ASSOCIATION OF EARLY ERCP AND DRAINAGE, MORTALITY, AND ICU ADMISSION

The association of early ERCP and drainage and outcomes is controversial and depends on AC severity. Hedjoudje et al[23] found that delayed ERCP (after 24 hours) is associated with ICU admission and mortality, in line with Huang et al[24], Hou et al[25], and Lavillegrand et al[26], who observed lower mortality when ERCP was conducted within 48 hours. Schwed et al[27] found no clinical effects of ERCP timing on adverse outcomes.

AC AND CONCURRENT ACUTE BILIARY PANCREATITIS

Liu et al[28] hypothesized that patients with both AC and concurrent acute biliary pancreatitis (ABP) are expected to exhibit a more pronounced inflammatory response than patients with AC alone and worse clinical outcomes. Then the authors conducted a single-center retrospective study with a two-year duration (358 patients were included). However, the study did not confirm their hypothesis because they found no significant impact of ABP on AC.

The authors were interested in biliary cholangitis and AC; other causes were excluded. The authors rely on the Tokyo criteria 13/18[29] to diagnose AC (systemic inflammation, cholestasis, and imaging). ABP in patients with AC is common and ranges from 18.7% and 23%[30]. We greatly acknowledge Liu et al’s effort despite the limitations of their study[28], including the retrospective design, single-center study, and not including albumin and C-reactive proteins, which significantly limited their results.

The current letter highlighted the importance of revising the Tokyo guidelines for AC to include RAR, NLR, PLR, diabetes, electrolyte imbalance, PAR, and C-reactive protein-to-albumin (CRPAR). In addition, concurrent acute pancreatitis is common, and one in five patients with AC is affected. Although the findings of Liu et al[28] suggested no association between concurrent acute pancreatitis in AC patients and ICU admission. However, the evidence is weak, and further studies on this common association are needed.

CONCLUSION

Old age, end-organ failure, RAR, NLR, PLR, the need for ventilator support, diabetes, electrolyte imbalance, PAR, CRPAR, GCS, and systolic hypertension are predictors of poor outcomes in AC of varying etiology, and concurrent acute pancreatitis was not associated with ICU admission.

ACKNOWLEDGEMENTS

We want to acknowledge the Saudi Digital Library for providing access to its databases. The authors also acknowledge Mohannad Osman for the recording of the audio core tip.