We read with outstanding interest the article by Celik et al.^[1] It is the first external description of the Hamburg’s preoperative pancreatic resection (PREPARE) score since it was published in 2014 by Uzunoglu et al.^[2] According to the latter authors, it is necessary to stratify risk of major morbidity and mortality before performing a pancreatic resection, and PREPARE score is a simple index, validated in a multicenter cohort of patients undergoing this group of technically demanding procedures.

 

With the aim of contributing to external validation of this score as well, we designed an observational prospective study immediately after publication of PREPARE score, carried out in our Hepato-pancreato-biliary Unit. Among all evaluated individuals with pancreatic or periampullary lesions (benign and malignant) at our institution, those patients between 18 and 80 years old who underwent major pancreatic resection (Whipple’s procedure as well as distal or total pancreatectomy) were included for this study. Exclusion criteria: patients with unresectable pancreatic or periampullary malignacies in whom operation was aborted or a palliative procedure without pancreatic resection was performed. After selection, 18 patients have been enrolled for this study from January 2015 to March 2016 in whom we calculated PREPARE score. A blood analysis was obtained immediately before operation but blood pressure and cardiac rate were registered upon hospital admission, the afternoon before surgery.

 

In our series, mean age was 63.8±8.1 years (range 45-75) and 15 patients were male. Two patients were American Society of Anesthesiologists (ASA) score I, 11 were ASA score II and 5 were ASA score III. Mean albumin was 3.5±0.5 g/dL (range 2.7-4.8); mean hemoglobin was 13.7±1.6 g/L (range 11.1-17.2); mean systolic blood pressure was 124.7±16.3 mmHg (range 98-155); mean cardiac rate was 76.7±13.8 beats/min (range 55-97). Whipple’s procedure was performed in 15 patients, while one individual underwent total pancreatectomy as well other 2 distal pancreatectomies (elective surgery in all cases). Two thirds of the individuals were diagnosed of pancreatic entity. According to these items, PREPARE score was calculated, obtaining a mean value of 6.8±3.9 points (range 0-14), classifying 8 individuals as low risk category, 5 as intermediate risk and 5 as high risk.

 

Major complications (Clavien ≥III) occurred in 5 patients (27.8%), with the following distribution: IIIa 5.6%, IIIb 5.6%, IV 11.1% and V 5.6%. Our only death corresponded to a female patient with Child-Pugh grade A liver cirrhosis due to hepatitis C virus, who developed progressive multiorgan failure without any evidence of technical complications. Pancreatectomy specific morbidities were: hemorrhage^[3] (grade A 5.6%, grade B 5.6%, grade C 11.1%), delayed gastric emptying^[4] (grade A 16.7%, grade B 11.1%) and fistula^[5] (grade A 5.6%, grade B 5.6%). Mean hospital stay was 16.6±11.2 days (range 6-47).

 

Statistical analysis (SPSS 18.0, IBM corporation) showed a significant higher proportion of severe postoperative complications (Clavien ≥III) in those patients included in the PREPARE intermediate risk group (2/5, 40.0%) and high risk group (3/5, 60.0%), though subjects classified as low risk did not suffer major morbidity (0%; P=0.02). Consequently, we obtained a relative risk of 0.200 (95% CI: 0.058-0.691) and 0.270 (95% CI: 0.104-0.716) for those patients with low risk versus those in the intermediate and high risk cohorts, respectively. However, we found no statistical differences in terms of postoperative hospital stay between the three groups (16.1 versus 18.6 versus 18.5 days, respectively).

 

The score proposed by Uzunoglu et al^[2] in 2014 seems to be easy-to-use, since it only includes 8 variables, and accurate (75.0%) as reported by the authors. To date, only the article by Celik et al^[1] has externally evaluated PREPARE score in a retrospective study, including patients operated from 2010 to 2015 in Turkey. Consequently, our preliminary study is the first prospective cohort of patients, specifically designed in order to validate this index, and we keep on recruiting patients. Currently, we have obtained statistical differences in terms of risk prediction for severe morbidity and mortality (0%, 40% and 60%, for low, intermediate and high risk groups, respectively) despite an extremely low number of patients. These figures are similar to those published when the score was described, excepting for the low risk cohort (18%, 40% and 66%, respectively), what might be explain due to our short series. Global major morbidity and mortality rates in our study do not differ much from the figures reported by Uzunoglu et al.^[2] On the contrary, PREPARE score does not statistically correlate with postoperative hospital stay in our series, what has not been previously evaluated.

 

We conclude that our preliminary results, as well as those reported by Celik et al,^[1] support the use of PREPARE score, though more large studies are still required. Hence, obtaining <6 points implies, in our series, a protective clinical context against major morbidity in those patients undergoing pancreatic resection, as defined for this index. Theoretically, the albumin value, which is weighted as the most important variable of this score (5 points), might be improved preoperatively by nutritional support. This reflects the capital importance of preoperative nutritional status as a factor associated with surgical complications after pancreatectomy.^[6]

 

 

Mario Rodriguez-Lopez, Martin Bailon-Cuadrado,

Francisco J Tejero-Pintor, Baltasar Perez-Saborido,

Enrique Asensio-Diaz and Asterio Barrera-Rebollo

 

Acknowledgments: Authors are thankful to all consultant surgeons from our HPB Surgery Unit, to the rest of surgeons and residents at the General and Digestive Surgery Department (Rio Hortega University Hospital, Valladolid, Spain) as well as to our OR and ward teams of nurses. Besides, RLM thanks his colleagues from the General, Visceral and Thoracic Surgery Department at the University Medical Center Hamburg-Eppendorf (leaded by Prof. Jakob R Izbicki), where he attended an Observership in 2014, and particularly to Dr. Matthias Reeh, who first provided the paper of the PREPARE score.

Disclosures: The results herein presented have been accepted for oral presentation at the Spanish National Surgical Congress, which takes place in Madrid (Spain) in November 2016. Authors declare that no financial support has been provided for this study.

 

1 Celik H, Kilic MO, Erdogan A, Ceylan C, Tez M. External validation of PREPARE score in Turkish patients who underwent pancreatic surgery. Hepatobiliary Pancreat Dis Int 2016;15:108-109. PMID: 26818552

2 Uzunoglu FG, Reeh M, Vettorazzi E, Ruschke T, Hannah P, Nentwich MF, et al. Preoperative pancreatic resection (PREPARE) score: a prospective multicenter-based morbidity risk score. Ann Surg 2014;260:857-864. PMID: 25243549

3 Wente MN, Veit JA, Bassi C, Dervenis C, Fingerhut A, Gouma DJ, et al. Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery 2007;142:20-25. PMID: 17629996

4 Wente MN, Bassi C, Dervenis C, Fingerhut A, Gouma DJ, Izbicki JR, et al. Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by the International Study Group of Pancreatic Surgery (ISGPS). Surgery 2007;142:761-768. PMID: 17981197

5 Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J, et al. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 2005;138:8-13. PMID: 16003309

6 Shirakawa H, Kinoshita T, Gotohda N, Takahashi S, Nakagohri T, Konishi M. Compliance with and effects of preoperative immunonutrition in patients undergoing pancreaticoduodenectomy. J Hepatobiliary Pancreat Sci 2012;19:249-258. PMID: 21667052

 

(doi: 10.1016/S1499-3872(16)60151-0)

Published online November 4, 2016.