Journal of Surgical Oncology 2014;110:720–726
Nonalcoholic Fatty Liver Disease After Pancreatoduodenectomy is Closely Associated With Postoperative Pancreatic Exocrine Insufficiency NAOYA NAKAGAWA, MD,* YOSHIAKI MURAKAMI, MD, KENICHIRO UEMURA, MD, TAKESHI SUDO, MD, YASUSHI HASHIMOTO, MD, NARU KONDO, MD, HAYATO SASAKI, MD, KEISUKE OKANO, MD, AND TAIJIRO SUEDA, MD Department of Surgery, Applied Life Sciences, Institute of Biochemical & Health Sciences, Hiroshima University, Hiroshima, Japan
Background: In recent years, nonalcoholic fatty liver disease (NAFLD) after pancreatoduodenectomy (PD) has become increasingly problematic. Our aims were to clarify the relationship between NAFLD and postoperative pancreatic exocrine function and to identify the risk factors for NAFLD after PD. Methods: Patients who underwent PD (n ¼ 104) were assessed with abdominal unenhanced computed tomography (CT) to determine the fatty liver changes and were given a 13C‐labeled mixed triglyceride breath test to measure pancreatic exocrine function. The percent 13CO2 cumulative dose at 7 hr (% dose 13C cum 7 hr) 150 g weekly) after PD, hepatitis virus infection and steroid therapy were also excluded from this analysis. As a result, this analysis included 104 of the 262 patients who underwent unenhanced CT for assessing the fatty changes of the liver and a 13C‐ labeled mixed triglyceride breath test (13C‐MTG‐BT) for assessing pancreatic exocrine function within 24 months after PD. The median follow‐up interval from surgery to these evaluations was 7.7 months, with a range of 2.5–23.6 months. Relationships between the occurrence of NAFLD and clinical factors including postoperative pancreatic exocrine function were analyzed by univariate and multivariate analyses.
Conflict of interest: None. *Correspondence to: Naoya Nakagawa, MD, Department of Surgery, Applied Life Sciences, Institute of Biochemical and Health Sciences, Hiroshima University, 1‐2‐3 Kasumi, Minami‐ku, Hiroshima 734‐8551, Japan. Fax: þ81‐82‐257‐5219. E‐mail: [email protected] Received 2 February 2014; Accepted 5 May 2014 DOI 10.1002/jso.23693 Published online 25 June 2014 in Wiley Online Library (wileyonlinelibrary.com).
NAFLD after Pancreatoduodenectomy
721
Surgical Procedures and Perioperative Management Most patients underwent pylorus‐preserving pancreatoduodenectomy (PPPD) [9,10]. PD with anterectomy was performed only if the tumor was close to the pylorus. Regional lymphadenectomy was performed only in malignancy or possibility of malignancy. To prevent severe postoperative diarrhea, dissection of the nerve plexus around the superior mesenteric artery (SMA) was avoided in most patients. However, when tumor invasion of the SMA nerve plexus was suspected, the right half of the SMA nerve plexus was dissected. Partial resection of the superior mesenteric vein or the portal vein (SMV/PV) was carried out if the surgeon observed tumor invasion to the SMV/PV at surgery. The pancreatic parenchyma was classified as having either a soft texture or a hard texture based upon intraoperative impressions of the texture of the remnant pancreas. Our method of pancreatoenteric anastomosis was a duct‐to‐mucosa PG [1]. The pancreatic juice flows through pancreaticogastric internal stent to the stomach. This stent ordinarily dropped spontaneously within 1 to 2 months after surgery. After pancreatic reconstruction, an end‐to‐side hepaticojejunostomy and an end‐to‐side duodenojejunostomy (or gastrojejunostomy) were performed with an antecolic Roux‐en Y type reconstruction. Postoperative adjuvant chemotherapy with gemcitabine and S‐1 was usually administered to patients with pancreatic carcinoma and biliary carcinoma as previously reported [11,12]. In brief, patients were treated with 10 cycles of gemcitabine plus S‐1 every 2 weeks. Each chemotherapy cycle consisted of intravenous gemcitabine (700 mg/ m2) on Day 1 and orally‐administered S‐1 (50 mg/m2) for 7 consecutive days, followed by a 1‐week pause. The body mass index (BMI) of each patient was recorded at every outpatient visit. The change in BMI was calculated by dividing the postoperative BMI value by the preoperative BMI value.
Evaluation of Nonalcoholic Fatty Liver Disease Computed tomography (CT) images were obtained with a 64‐ multidetector CT scanner (Light Speed VCT, GE, Milwaukee, WI) without intravenous contrast medium during a single breathhold. Scanning parameters for CT were as follows: beam collimation 40 mm, table speed 55 mm/rotation, gantry rotation time 0.4 sec, 120 kVp, 100– 750 mA. The raw data set was reconstructed at 5‐mm thickness. The CT images were reviewed by an independent radiologist blinded to the patients’ clinical features. For each patient, the average CT attenuation values of six regions in the liver and two regions in the spleen were measured. Each region of interest was a circular area with a diameter of 20 mm (Fig. 1). The mean liver‐to‐spleen attenuation ratio (L/S ratio) on CT was calculated, and the postoperative NAFLD was defined as an L/S ratio of less than 0.9 or a liver attenuation value of at least 10 Hounsfield units (HU) lower than the splenic attenuation value [13,14]. The CT attenuation value and L/S ratio were examined preoperatively and postoperatively.
Pancreatic Endocrine and Exocrine Function Pancreatic endocrine function was analyzed by measuring serum levels of hemoglobin A1c according to the National Glycohemoglobin Standardization Program (NGSP‐HbA1c). Serum NGSP‐HbA1c levels were determined preoperatively and at the time of the postoperative CT to evaluate the fatty changes of the liver. Impaired pancreatic endocrine function was diagnosed if serum NGSP‐HbA1c levels exceeded 6.5% or if the patient needed diabetes therapy (oral hypoglycemic agent or insulin) [15,16]. Pancreatic exocrine function was evaluated with a 13C‐labeled mixed triglyceride breath test. A description of the 13C‐labeled mixed triglyceride breath test was published previously [8,17,18]. The breath test was analyzed in all 104 patients who could tolerate a Journal of Surgical Oncology
Fig. 1. A 69‐year‐old woman underwent pancreatoduodenectomy for pancreatic cancer. The computed tomography (CT) scan shows the areas used to calculate the mean liver CT attenuation and the liver‐to‐spleen attenuation ratio. The average CT attenuation values of 6 regions in the liver and 2 regions in the spleen were measured. Each region of interest was a circular area with a diameter of 20 mm. Postoperative CT of this case shows typical nonalcoholic fatty liver disease (the mean liver CT attenuation value: 12.9 HU, L/S ratio: 0.29) 6 months after surgery. normal solid diet, and was performed at the time of the postoperative CT evaluation for NAFLD. In brief, if patients received pancreatic enzyme replacement therapy, the oral pancreatic enzyme supplement was stopped 3 days before the day of the breath test. The test meal consisted of 20 g of fat and 200 mg of 13C‐labeled mixed triglycerides consisting of naturally occurring long‐chain fatty acids (Chlorella Industry Co., LTD, Tokyo, Japan). Enrichment of 13CO2/12CO2 in collected breath samples was measured by infrared spectrophotometry with an Otsuka Ubit‐IR 300 (Otsuka Electronics Co. Ltd., Osaka, Japan), and the results were expressed in accordance with the Pee Dee Belemnite international standard, as described previously [19,20]. Cumulative percentages of label recovery (% dose 13C cum) were calculated with the trapezoidal rule. From these data, the following parameter of fat assimilation was derived: the maximum percentage of the administered dose of 13C excreted per hours (% dose 13C cum 7 hr), which was used to assess pancreatic exocrine function; values less than 5% were considered diagnostic for PEI [8,17].
Measurement of Pancreatic Parenchymal Thickness The diameter of the main pancreatic duct (MPD) and the horizontal thickness of the pancreatic parenchyma were measured on the CT images, and the pancreatic parenchymal thickness was calculated by subtracting the diameter of the MPD from the horizontal thickness of the pancreas [21,22]. Preoperatively, the diameter of the MPD was measured along the presumed pancreatic resection line, which usually ran through the body of the pancreas anterior to the aorta or portal vein [23]. Postoperatively, the maximum caliber of pancreatic duct and the anterior–posterior width of the entire gland were measured in the pancreas wherever the greatest length of duct could be identified [24].
Statistical Analysis Continuous variables were expressed as mean standard deviation. Categorical data were compared with the x2 test or Fisher’s exact test as appropriately. The paired t‐test or Student’s t‐test was used for comparison of continuous variables between the two groups. Factors
722
Nakagawa et al.
found to be significant on univariate analysis were subjected to multivariate analysis with a multiple logistic regression model. Significance was defined at a P value of
Nonalcoholic Fatty Liver Disease After Pancreatoduodenectomy is Closely Associated With Postoperative Pancreatic Exocrine Insufficiency NAOYA NAKAGAWA, MD,* YOSHIAKI MURAKAMI, MD, KENICHIRO UEMURA, MD, TAKESHI SUDO, MD, YASUSHI HASHIMOTO, MD, NARU KONDO, MD, HAYATO SASAKI, MD, KEISUKE OKANO, MD, AND TAIJIRO SUEDA, MD Department of Surgery, Applied Life Sciences, Institute of Biochemical & Health Sciences, Hiroshima University, Hiroshima, Japan
Background: In recent years, nonalcoholic fatty liver disease (NAFLD) after pancreatoduodenectomy (PD) has become increasingly problematic. Our aims were to clarify the relationship between NAFLD and postoperative pancreatic exocrine function and to identify the risk factors for NAFLD after PD. Methods: Patients who underwent PD (n ¼ 104) were assessed with abdominal unenhanced computed tomography (CT) to determine the fatty liver changes and were given a 13C‐labeled mixed triglyceride breath test to measure pancreatic exocrine function. The percent 13CO2 cumulative dose at 7 hr (% dose 13C cum 7 hr) 150 g weekly) after PD, hepatitis virus infection and steroid therapy were also excluded from this analysis. As a result, this analysis included 104 of the 262 patients who underwent unenhanced CT for assessing the fatty changes of the liver and a 13C‐ labeled mixed triglyceride breath test (13C‐MTG‐BT) for assessing pancreatic exocrine function within 24 months after PD. The median follow‐up interval from surgery to these evaluations was 7.7 months, with a range of 2.5–23.6 months. Relationships between the occurrence of NAFLD and clinical factors including postoperative pancreatic exocrine function were analyzed by univariate and multivariate analyses.
Conflict of interest: None. *Correspondence to: Naoya Nakagawa, MD, Department of Surgery, Applied Life Sciences, Institute of Biochemical and Health Sciences, Hiroshima University, 1‐2‐3 Kasumi, Minami‐ku, Hiroshima 734‐8551, Japan. Fax: þ81‐82‐257‐5219. E‐mail: [email protected] Received 2 February 2014; Accepted 5 May 2014 DOI 10.1002/jso.23693 Published online 25 June 2014 in Wiley Online Library (wileyonlinelibrary.com).
NAFLD after Pancreatoduodenectomy
721
Surgical Procedures and Perioperative Management Most patients underwent pylorus‐preserving pancreatoduodenectomy (PPPD) [9,10]. PD with anterectomy was performed only if the tumor was close to the pylorus. Regional lymphadenectomy was performed only in malignancy or possibility of malignancy. To prevent severe postoperative diarrhea, dissection of the nerve plexus around the superior mesenteric artery (SMA) was avoided in most patients. However, when tumor invasion of the SMA nerve plexus was suspected, the right half of the SMA nerve plexus was dissected. Partial resection of the superior mesenteric vein or the portal vein (SMV/PV) was carried out if the surgeon observed tumor invasion to the SMV/PV at surgery. The pancreatic parenchyma was classified as having either a soft texture or a hard texture based upon intraoperative impressions of the texture of the remnant pancreas. Our method of pancreatoenteric anastomosis was a duct‐to‐mucosa PG [1]. The pancreatic juice flows through pancreaticogastric internal stent to the stomach. This stent ordinarily dropped spontaneously within 1 to 2 months after surgery. After pancreatic reconstruction, an end‐to‐side hepaticojejunostomy and an end‐to‐side duodenojejunostomy (or gastrojejunostomy) were performed with an antecolic Roux‐en Y type reconstruction. Postoperative adjuvant chemotherapy with gemcitabine and S‐1 was usually administered to patients with pancreatic carcinoma and biliary carcinoma as previously reported [11,12]. In brief, patients were treated with 10 cycles of gemcitabine plus S‐1 every 2 weeks. Each chemotherapy cycle consisted of intravenous gemcitabine (700 mg/ m2) on Day 1 and orally‐administered S‐1 (50 mg/m2) for 7 consecutive days, followed by a 1‐week pause. The body mass index (BMI) of each patient was recorded at every outpatient visit. The change in BMI was calculated by dividing the postoperative BMI value by the preoperative BMI value.
Evaluation of Nonalcoholic Fatty Liver Disease Computed tomography (CT) images were obtained with a 64‐ multidetector CT scanner (Light Speed VCT, GE, Milwaukee, WI) without intravenous contrast medium during a single breathhold. Scanning parameters for CT were as follows: beam collimation 40 mm, table speed 55 mm/rotation, gantry rotation time 0.4 sec, 120 kVp, 100– 750 mA. The raw data set was reconstructed at 5‐mm thickness. The CT images were reviewed by an independent radiologist blinded to the patients’ clinical features. For each patient, the average CT attenuation values of six regions in the liver and two regions in the spleen were measured. Each region of interest was a circular area with a diameter of 20 mm (Fig. 1). The mean liver‐to‐spleen attenuation ratio (L/S ratio) on CT was calculated, and the postoperative NAFLD was defined as an L/S ratio of less than 0.9 or a liver attenuation value of at least 10 Hounsfield units (HU) lower than the splenic attenuation value [13,14]. The CT attenuation value and L/S ratio were examined preoperatively and postoperatively.
Pancreatic Endocrine and Exocrine Function Pancreatic endocrine function was analyzed by measuring serum levels of hemoglobin A1c according to the National Glycohemoglobin Standardization Program (NGSP‐HbA1c). Serum NGSP‐HbA1c levels were determined preoperatively and at the time of the postoperative CT to evaluate the fatty changes of the liver. Impaired pancreatic endocrine function was diagnosed if serum NGSP‐HbA1c levels exceeded 6.5% or if the patient needed diabetes therapy (oral hypoglycemic agent or insulin) [15,16]. Pancreatic exocrine function was evaluated with a 13C‐labeled mixed triglyceride breath test. A description of the 13C‐labeled mixed triglyceride breath test was published previously [8,17,18]. The breath test was analyzed in all 104 patients who could tolerate a Journal of Surgical Oncology
Fig. 1. A 69‐year‐old woman underwent pancreatoduodenectomy for pancreatic cancer. The computed tomography (CT) scan shows the areas used to calculate the mean liver CT attenuation and the liver‐to‐spleen attenuation ratio. The average CT attenuation values of 6 regions in the liver and 2 regions in the spleen were measured. Each region of interest was a circular area with a diameter of 20 mm. Postoperative CT of this case shows typical nonalcoholic fatty liver disease (the mean liver CT attenuation value: 12.9 HU, L/S ratio: 0.29) 6 months after surgery. normal solid diet, and was performed at the time of the postoperative CT evaluation for NAFLD. In brief, if patients received pancreatic enzyme replacement therapy, the oral pancreatic enzyme supplement was stopped 3 days before the day of the breath test. The test meal consisted of 20 g of fat and 200 mg of 13C‐labeled mixed triglycerides consisting of naturally occurring long‐chain fatty acids (Chlorella Industry Co., LTD, Tokyo, Japan). Enrichment of 13CO2/12CO2 in collected breath samples was measured by infrared spectrophotometry with an Otsuka Ubit‐IR 300 (Otsuka Electronics Co. Ltd., Osaka, Japan), and the results were expressed in accordance with the Pee Dee Belemnite international standard, as described previously [19,20]. Cumulative percentages of label recovery (% dose 13C cum) were calculated with the trapezoidal rule. From these data, the following parameter of fat assimilation was derived: the maximum percentage of the administered dose of 13C excreted per hours (% dose 13C cum 7 hr), which was used to assess pancreatic exocrine function; values less than 5% were considered diagnostic for PEI [8,17].
Measurement of Pancreatic Parenchymal Thickness The diameter of the main pancreatic duct (MPD) and the horizontal thickness of the pancreatic parenchyma were measured on the CT images, and the pancreatic parenchymal thickness was calculated by subtracting the diameter of the MPD from the horizontal thickness of the pancreas [21,22]. Preoperatively, the diameter of the MPD was measured along the presumed pancreatic resection line, which usually ran through the body of the pancreas anterior to the aorta or portal vein [23]. Postoperatively, the maximum caliber of pancreatic duct and the anterior–posterior width of the entire gland were measured in the pancreas wherever the greatest length of duct could be identified [24].
Statistical Analysis Continuous variables were expressed as mean standard deviation. Categorical data were compared with the x2 test or Fisher’s exact test as appropriately. The paired t‐test or Student’s t‐test was used for comparison of continuous variables between the two groups. Factors
722
Nakagawa et al.
found to be significant on univariate analysis were subjected to multivariate analysis with a multiple logistic regression model. Significance was defined at a P value of
