This study was conducted to analyze the protective effect of duodenal preservation in decreasing the incidence of hepatic steatosis after pancreatectomy. The incidence of hepatic steatosis post PD varies between 15.6%-66%[5, 15–17], and the majority of these studies are from East Asia. The incidence of hepatic steatosis post DPPHR is reported only in two studies: Kato et al. [9] noticed an incidence of 10%, while Chao et al. [18] observed an incidence of 4% in his cohort of 25 patients post laparoscopic DPPHR. In our study, post-op hepatic steatosis was seen in 17.5% and 37.5%, respectively in the DPHHR and PD group respectively.
Multiple studies, both randomized and non-randomized, comparing PD vs. duodenum preserving head resection have been conducted in the past for variable outcomes. While certain studies like Pedrazolli et al. [19], Witzigmann et al. [20], Sun et al[21] compared post-op symptoms like pain-relief and quality of life parameters (QOL); Farkas et al[22], Klempa et al. [23] and Izbicki et al. [24] compared long term morbidities like post-pancreatectomy new-onset DM (pODM), exocrine insufficiency and post-op occupational rehabilitation. A study by Kato et al. [9] was one with the primary aim of comparing post-op NAFLD, cholangitis and liver fibrosis between DPPHR vs PD.
The intra-operative texture of the pancreas was firm in 92.5% and 37.5% in the DPPHR and PD groups, respectively. It did not significantly predict post-op hepatic steatosis in our cohort. As PEI (pancreatic exocrine insufficiency) is a risk factor for post-op hepatic steatosis, multiple studies [25, 26] showed an association between pancreatic texture and PEI and hence inferred its association with post-op hepatic steatosis also, while no direct association tests were done. Tsunematsu et al. [27] showed a significant association between hard pancreatic texture and post-op hepatic steatosis.
Post-operative clinical pancreatic exocrine insufficiency (PEI) was seen in 35% and 15% in the DPPHR and PD group respectively. It was not a significant predictor of post-op hepatic steatosis in our cohort. In the study reported by Kato et al[9], pancreatic exocrine insufficiency was not a significant predictor of post-op hepatic steatosis in univariate analysis. Tsunematsu et al. [27] showed pancreatic exocrine index (product of intra-operative amylase activity of pancreatic juice and residual pancreatic volume after surgery) significantly associated with post-op hepatic steatosis in multivariate analyses.
Post-op Diabetes mellitus was seen in 50% and 47.5% in the DPPHR and PD group, respectively. It was not a significant predictor of post-op hepatic steatosis. Parth Shah et al. [28], in their systematic review, enlist post-operative diabetes as a risk factor for hepatic steatosis. In a study by Tsunematsu et al. (27), the presence of Diabetes mellitus at a follow-up of 1 year was not significantly associated with post-op hepatic steatosis. The mechanism of hepatic steatosis after pancreatectomy, caused by insufficient levels of glucoregulatory hormones in the setting of a pancreatectomy, can be described as pancreatogenic diabetes (type 3C), which is like a type 1 than type 2 diabetes. As type 1 DM, in turn, is less commonly associated with hepatic steatosis, this could explain the lack of association between post-op DM and hepatic steatosis.
The cause for pancreatectomy was also found to be related to the emergence of hepatic steatosis. Patients with malignancies have greater rates of malnutrition due to a variety of comorbidities and disease processes contributing to the incidence of hepatic steatosis.[16] In our study, pre-op diagnosis (benign vs malignant) was not a significant predictor of post-op hepatic steatosis. Tanaka et al. [29] showed pancreatic head cancer as a significant risk factor for hepatic steatosis postoperatively on multivariate logistic regression analysis. As per the prognostic nutrition index, the majority were seriously malnourished in both the malignant and benign aetiologies in our cohort, thus explaining the lack of influence of pre-operative diagnosis on post-operative hepatic steatosis.
In our cohort, PD was a significant risk factor for post-op hepatic steatosis (OR: 2.83, CI: 1.003–7.977), and the risk was seen to persist after propensity score matching. There was a significant worsening of pre-op CT attenuation values in the PD (40%) vs DPPHR (25%) group (P = 0.41). No significant difference in NAFLD fibrosis score was noted between the two groups. Kato et al. [9] also reported on the role of preserving duodenum in preventing post-op hepatic steatosis. In the case of DPPHR, Chao et al. [18] reported a further decrease in the incidence of post-op hepatic steatosis with minimally invasive techniques (laparoscopy). However, no significant difference was seen when we compared the incidence of post-operative hepatic steatosis between minimally invasive (robot-assisted and laparoscopic) vs open DPPHR.
This study had a few drawbacks: it was a single-center study with moderate volume and could only be regarded as an exploratory investigation. Consequently, a large multicentre investigation would be ideal to show that the findings of this study may be validated.
In conclusion, duodenal resection surgeries have more association with hepatic steatosis compared to duodenal preserving surgeries, independent of the effects of post-op exocrine insufficiency and Diabetes mellitus.