급성췌장염에서 애틀랜타 분류 개정안 적용과 임상경과 예측: 후향적, 다기관 연구
Revised Atlanta Classification of Acute Pancreatitis Can Predict Clinical Outcome Better: a Retrospective, Multicenter Study
Article information
Abstract
배경/목적:
2012년 급성췌장염의 애틀랜타 분류 개정안이 발표되었고 이후 널리 이용하고 있다. 본 연구는 1992년 애틀랜타 분류와 2012년 개정안을 이용하여 급성췌장염 환자를 분류한 후 임상경과를 비교하였다.
방법:
2012년 1월부터 2013년 7월 사이 대구경북지역 6개 대형 병원에 입원한 총 574명의 급성췌장염 환자를 대상으로 후향적 연구를 시행하였다. 1992년 애틀랜타 분류와 2012년 개정안을 이용하여 중등도 평가를 시행하고 인구 통계 자료, 장기부전, 국소합병증, 입원기간, 임상경과 등을 분석하였다.
결과:
대상 중 377명(65.7%)이 남성이었다. 연령의 중간값은 55.4세였다. 급성췌장염의 2대 원인은 음주(238명, 41.5%)와 담석(193명, 33.6%)이었다. 2012년 개정안에 따르면 경증췌장염은 356명(62%), 중등췌장염은 197명(34.3%), 중증췌장염은 21명(3.7%)이었다. 입원기간은 중증도 증가에 따라 통계학적으로 유의하여 증가하였다(경증 5.9일, 중등 8.3일, 중증 13일, p<0.001). 모든 경증 및 중등췌장염 환자는 임상적으로 호전되었다. 하지만, 호전을 보이지 않은 11명은 모두 중증췌장염에 속했다.
결론:
애틀랜타 분류 개정안이 급성췌장염의 임상경과를 예측하는 데 있어 좋은 방법이다.
Trans Abstract
Background/Aims:
The 2012 revision of the Atlanta classification of acute pancreatitis (AP) by international consensus has been published and in use. This study investigated and compared clinical outcome of patients with AP stratified according to the 1992 Atlanta classification and revised classification.
Methods:
A total of 574 AP patients from six referral hospitals between January 2012 and July 2013 were included. Medical records were reviewed retrospectively. Severity assessment according to both classifications was done. Demographics, organ failure, local complications, length of stay, and clinical outcome were recorded.
Results:
There were 377 males (65.7%). Median age was 55.4 years. Two most common causes of AP were alcohol (n=238, 41.5%) and gallstone (n=193, 33.6%). According to revised classification, there were mild (n=356, 62%), moderately severe (n=197, 34.3%), and severe AP (n=21, 3.7%). Length of stay showed gradual increment with increase in degrees of severity according to the revised classification (5.9 days in mild AP, 8.3 days in moderately severe AP, and 13 days in severe AP, p<0.001). All the patients with mild and moderately severe AP improved, but all the 11 cases without improvement belonged to severe AP.
Conclusions:
The revised classification seems to be a good predictor for clinical outcome of AP.
INTRODUCTION
The original Atlanta classification of acute pancreatitis (AP) of 1992 has been widely used since its publication [1], but has been criticized for confusing definitions and difficulty in predicting clinical outcome [2]. The 2012 revision of the Atlanta classification and definitions by international consensus has been published and in use [3]. The 1992 Atlanta classification divided AP into two groups: mild and severe AP. Definition of severe AP included presence of local complications on imaging studies, organ failure, and/or acute physiology and chronic health examination evaluation II (APACHE-II) score of 8 or greater or Ranson score of 3 and greater [1]. On the other hand, the 2012 revision divided AP into three groups: mild, moderately severe, and severe AP. Moderately severe AP is defined as presence of local complications or co-morbidities with organ failure that resolved within 48 hours. Severe AP is defined as presence of persistent organ failure. Definition of organ failure is different between two classification. The 2012 revision used the modified Marshall scoring system [3,4] while the 1992 Atlanta classification defined organ failure as presence of any of the followings: systolic blood pressure less than 90 mmHg, PaO2 less than 60 mmHg, serum creatinine 2 mg/dL and greater, gastrointestinal bleeding greater than 500 mL per 24 hours[1].
The aim of this study is to investigate and compare the clinical outcome of patients with AP stratified according to the 1992 Atlanta classification and revised classification.
METHODS
1. Patients
Patients admitted with AP between January 2012 and June 2013 were included. Six referral hospitals in the Daegu-Gyungbuk area participated in this study. Only patients who were older than 18 years of age were included. Patients with acute exacerbation of chronic pancreatitis, established endstage renal disease, acute pancreatitis caused by pancreatic cancer, ampulla of Vater tumor, and intraductal papillary mucinous neoplasm of the pancreas, and post-endoscopic retrograde cholangiopancreatography pancreatitis were excluded. Medical records of the patients were retrospectively reviewed for etiology, severity according to the 1992 Atlanta classification and the revised classification, clinical features, treatment modality, length of hospital stay, presence of organ failure, and clinical outcome.
2. Definitions
AP was diagnosed when at least two of the following three criteria were present: (1) abdominal pain consistent with acute pancreatitis; (2) serum amylase or lipase level at least three times greater than the upper limit of normal; and (3) characteristic findings compatible with acute pancreatitis on contrast-enhanced computed tomography, magnetic resonance imaging or transabdominal ultrasonography [3]. According to the 1992 Atlanta classification, AP was stratified into mild AP and severe AP [1]. In the revised Atlanta classification, it was stratified into mild AP, moderately severe AP, and severe AP [3]. Patients with moderately severe AP had transient organ failure that resolved within 48 hours, local or systemic complications without persistent organ failure. Organ failure was defined as a score of 2 or more for one of three organ systems using the modified Marshall scoring system [4]. Three organ systems included respiratory, cardiovascular, and renal systems. Local complications included pancreatic and peripancreatic collections, gastric outlet dysfunction, splenic and portal vein thrombosis, and colonic necrosis [3]. A systemic complication was defined as exacerbation of pre-existing co-morbidity, such as coronary artery disease or chronic obstructive pulmonary disease, which had been precipitated by the attack of acute pancreatitis. Clinical outcome was divided into improvement and no improvement. Improvement was defined as resolution of systemic complication with or without pancreatic pseudocyst or walled-off necrosis. No improvement was defined as persistent systemic complication or death from AP. Definitions of various features of AP (e.g., interstitial edematous pancreatitis, necrotizing pancreatitis, acute peripancreatic fluid collection, pancreatic pseudocyst, acute necrotic collection, walled-off necrosis) followed those described in Banks et al. [3]. Systemic inflammatory response syndrome (SIRS) was defined as previously described by Muckart et al. [5]. It was determined at 24 hours of admission.
3. Statistical analysis
SPSS 19.0 software for Windows (IBM, Armonk, NY, USA) was used for statistical analysis. Data were presented as number (%) and median (interquartile range) for continuous variables. The Mann-Whitney U test or Kruskal-Wallis test was used to compare the continuous variables for more than two groups. A p value <0.05 was considered statistically significant.
RESULTS
1. Patient characteristics
A total of 574 patients were included in this study (Table 1). Their median age was 55.4 years (interquartile range 44-71 years). There were 377 male patients (65.7%). The etiologies of AP were as follows in decreasing order: alcohol, gallstone, idiopathic, hypertriglyceridemia, and pancreas divisum. There were 528 patients (92%) with interstitial edematous pancreatitis and 46 (8%) with necrotizing pancreatitis. There were 356 patients (62.0%) with mild AP, 197 (34.3%) with moderately severe AP, and 21 (3.7%) with severe AP.
2. Complications and treatment modalities
Pancreatic infection occurred in 34 patients (5.9%) (Table 1). Respiratory tract infection occurred in 41 patients (7.1%) and urinary tract infection in 27 (4.7%). Seven patients (1.2%) required ventilator care and four patients (0.7%) underwent renal replacement therapy. Seventy-one patients (12.4%) had nutritional support. Types of nutritional support included enteral nutrition via nasojejunal tube in 17 patients (3.1%), total parenteral nutrition in 52 (9.6%), and both in 2 (0.4%).
Drainage for AP-related complications was done in 18 patients (3.1%). Types of drainage included percutaneous drainage in 2 patients (1.8%), endoscopic transpapillary drainage in 7 (6.1%), endoscopic transmural drainage of pancreatic pseudocyst in 3 (2.6%), percutaneous abscess drainage in 1 (0.9%) and combination of more than two methods in 5 (4.3%). Surgical management of AP-related complications was done in 3 patients (0.5%).
Organ failure was present in 61 patients (10.6%). SIRS was present in 306 patients (53.3%). About one third of local complications developed within the first 4 weeks after onset of AP. Acute peripancreatic fluid collection was found in 213 patients (37.1%). Acute necrotic collection was found in 42 patients (7.3%). Its location was intrapancreatic alone in 21 (3.7%), extrapancreatic alone in 8 (1.4%), and both in 13 (2.3%). One hundred and four patients (18.2%) had local complications that occurred more than 4 weeks after the onset of AP. Pancreatic pseudocyst was found in 95 patients (16.7%) and walled-off necrosis in 64 (11.2%). Both pancreatic pseudocyst and walled-off necrosis were found in 57 patients. Systemic complications occurred in 26 patients (4.5%).
3. Severity assessment
According to the 1992 Atlanta classification, there were 320 patients (55.7%) with mild AP and 254 (44.3%) with severe AP (Table 2). According to the revised classification, there were 356 (62%) patients with mild AP, 197 (34.3%) with moderately severe AP, and 21 (3.7%) with severe AP.
4. Comparison between two classifications
In terms of the clinical outcomes, all of the patients with mild AP according to the 1992 Atlanta classification showed improvement (Table 3). But, 11 patients (5.4%) with severe AP according to the 1992 Atlanta classification showed no improvement. They comprised 5.4% of the patients with severe AP. All of the patients with mild AP and moderately severe AP according to the revised classification showed improvement. However, those 11 patients who showed no improvement comprised 64.7% of the patients with severe AP according to the revised classification (Table 4). Death occurred in 10 patients (1.7%); all the patients had severe AP according to both classifications.
When length of stay was compared according to the severity and classification, patients with severe AP according to both classifications had significantly longer stay (Table 5). But, length of stay showed gradual increment with increase in degrees of severity according to the revised classification (5.9 days in mild AP, 8.3 days in moderately severe AP, and 13 days in severe AP, p-value <0.001).
DISCUSSION
According to the revised Atlanta classification, proportions of mild AP, moderately severe AP and severe AP were 62%, 34.3% and 3.7%, respectively. On the other hand, those of mild AP and severe AP according to the 1992 Atlanta classification were 55.7% and 44.3%, respectively. While there was no patient with severe AP as defined in the revised classification who satisfied definition of mild AP in the 1992 Atlanta classification, there were 39.4% and 52.4% of patients with severe AP as defined in the 1992 Atlanta classification who satisfied definitions of mild and moderately severe AP in the revised classification. All of the patients with mild and moderately severe AP as defined in the revised classification and all of the patients with mild AP as defined in the 1992 Atlanta classification showed improvement. Those 11 patients with no improvement belonged to severe AP as defined in the revised classification. In addition, clinical outcome and length of stay correlated with severity in revised classification.
Severe AP according to the 1992 Atlanta classification is defined as presence of local complications and/or presence of organ failure and/or APACHE II score 8 or greater or Ranson score 3 or greater [1]. This broad definition of severe AP results in a heterogeneous group of patients with varying degrees of severity and significantly different mortality rates [2]. According to a retrospective study of 207 patients admitted with severe AP, patients with no organ failure had significantly shorter length of hospital stay, less need for intensive care unit care, shorter stay in the intensive care unit, and decreased mortality [2]. Since patients with severe AP and no organ failure showed lower mortality when compared with those with severe AP and organ failure (2% vs. 46%, p<0.01), Vege et al. [2] proposed that revision of the 1992 Atlanta classification should include this particular group of patients with severe AP and no organ failure as a patient group of “moderately severe AP”.
Since the publication of the revised classification, multiple studies have evaluated the revised classification [6-9]. All of the studies validated the clinical utility of the revised classification regardless of the study design.
A similar study of 553 patients from single center in Korea was previously published [7]. Although this study included a smaller number of patients with AP, this is a multicenter study from a specific region of Korea. Therefore, the results of this study may be applied to everyday clinical practice. Similar to the findings of this study, Choi et al. [7] reported that there were statistically significant differences regarding the need for intervention, duration of hospital stay, need for ICU care, or in-hospital mortality among mild, moderately severe, and severe AP.
There are some limitations regarding this study. First of all, this is a retrospective study. Nevertheless, our data from multiple teaching hospitals revealed valuable additional information on clinical features of AP in Korean population. And the results of this study would provide a fertile ground for future prospective studies on AP in Korean population. Secondly, the management of AP was not uniform since multiple centers with multiple gastroenterologists participated. Only 12.4% of the patients received nutritional support. Since there is no universally accepted management guideline and a recently published Korean guideline [10] needs to be updated, there is a room for improvements regarding management of AP in near future. Thirdly, clinical outcomes of the patients with infected necrosis were not analyzed. Previous studies reported that the patients with infected necrosis have worse clinical outcome and markedly increased mortality rate [7,9]. Because this study was not designed to evaluate the patients with infected necrosis, this particular subgroup of patients should be evaluated in a larger, prospective study.
In conclusion, the revised classification seems to be a good predictor for clinical outcome of AP. A prospective, multicenter study would be helpful in validation of these findings.
Notes
Conflict of Interest
All authors disclosed no financial relationships relevant to this publication.
Previous scientific presentations
Some of the data have been presented in 63rd Congress of the Korean Society of Gastrointestinal Endoscopy on November 23, 2013.