Benjamin Darnis

Porcine models for the study of small-for-size syndrome and portal inflow modulation: literature review and proposal for a standardized nomenclature: Porcine models for the study of small-for-size syndrome and portal inflow modulation: literature review and proposal for a standardized nomenclature

Porcine models of extended hepatectomy and liver transplantation (LT) of reduced graft have been widely used for studying the small‐for‐size (SFS) syndrome and the various modalities of portal inflow modulation (PIM). However, considerable heterogeneity exists among the studies and their results. The aim of this review was to assess the main advantages and drawbacks of the different porcine models of SFS LT and SFS hepatectomy, and propose a standardized anatomical nomenclature for the various models. The MEDLINE database was searched for articles reporting porcine models of reduced graft LT or hepatectomy of more than 65%. Nineteen articles on SFS LT matched our inclusion criteria, including 10 articles reporting a model of PIM. Twenty‐seven articles reporting a model of posthepatectomy SFS were identified, of which 16 reported a model of PIM. Subtotal hepatectomy (i.e. resection of all segments except segment 1) without inflow occlusion, left trisectionectomy with inflow occlusion, and LT of a right lateral section including the caudate lobe in a larger recipient appeared to be the most suitable porcine models for studying the SFS syndrome. All three models were appropriate for assessing the surgical and pharmaceutical PIM modalities, except for those involving the splenic flow.

Hepatic Hemodynamic Changes After Liver Resection: a Reflection of the Complex Relationship Between Portal Vein Flow, Hepatic Artery Flow and Portal Pressure

To the Editor, We read with great interest the article by Golriz et al. in which the authors studied the hepatic hemodynamic consequences of stepwise hepatectomy in swine. According to their findings, stepwise hepatectomy of 25, 50 and 75% resulted in a gradual decrease of hepatic artery flow normalized to liver remnant volume (HAF/LRV) and a gradual increase of portal vein pressure (PVP) and portal vein flow normalized to liver remnant volume (PVF/LRV). The authors concluded that the compensatory mechanism interacting between HAF/LRV and PVF/LRV led to a decrease of oxygenated hepatic blood inflow combined with an increase of inflow pressure, which may represent one of the main mechanisms involved in posthepatectomy liver failure following extended liver resection. This work underlines two of the main mechanisms involved in the small-for-size syndrome (SFSS): portal hyperperfusion and the hepatic artery buffer response, which consists of an inverse relationship between HAF/LRV and PVF/LRV. , 3 Both mechanisms have been mainly described after partial liver transplantation, but also in several experimental studies of partial hepatectomy in swine. –7 To address these issues, the authors insightfully raised the fact that potential means to decrease portal flow may improve the outcome of extended liver resection. We would like to emphasize that the large experience gained from partial liver transplantation has allowed developing several techniques of portal inflow modulation, such as splenic artery ligation or embolization, splenectomy or porto-caval shunts, some of which have already been proposed after liver resection in both experimental –9 and clinical settings. , 11 Moreover, pharmaceutical modalities of portal inflow modulation such as the use of somatostatin infusion may represent a promising alternative to surgical techniques, having the advantage of being non-invasive and reversible. Besides, we were astonished by the fact that the authors reported an overall 33 % increase of PVP after stepwise hepatectomy, since this was not consistent with our previous findings. , 12 Indeed, we recently demonstrated that 70 % hepatectomy in swine did not induce any significant acute portal hypertension and that the latter occurred only after 90 % hepatectomy, which comprised the additional excision of right lateral lobe. The main reason for these discrepant results may be related to the fact that the authors did not consider the hepatic venous pressure gradient (HVPG) to assess the variations of portal pressure, but only referred to PVP gross value. We believe that HVPG (the difference between the portal vein and the hepatic vein pressure [HVP]) is a much more reliable measurement of portal hypertension compared to PVP alone, since PVP alone is not only influenced by the pressure within the hepatic sinusoids but also depends on central venous pressure and systemic hemodynamics. Of note, in the study byGolriz et al., the mean arterial pressure dropped from 100 to 70 mmHg after 75 % hepatectomy, and even after 5 min, it only reached 80 mmHg, which was still significantly below the baseline value before completing 75 % hepatectomy, suggesting that hemodynamic conditions were not perfectly stable. More importantly, if HPV values had been used to calculate HVPG, a value of 9.4 mmHg would have been found after 50 % hepatectomy (HVPG = PVP–HPV = 15– * Kayvan Mohkam kayvan.mohkam@chu-lyon.fr

Respiratory depression related to multiple drug–drug interactions precipitated by a fluconazole loading dose in a patient treated with oxycodone

We present a case of respiratory depression likely due to oxycodone and precipitated by fluconazole. Oxycodone demethylation is catalyzed by CYP3A4 and CYP2D6 [1], with contribution ratios of 0.54 and 0.2, respectively [2]. Coadministration of oxycodone with strong CYP3A4 inhibitors results in raised oxycodone plasma levels [3–5]. Fluconazole inhibits CYP2C9 and CYP2C19 and is also a moderate inhibitor of CYP3A4 [6, 7]. A 51-year-old man (89 kg) with a history of chronic pain after seven surgical procedures for Scheuermann’s disease was admitted for pressure ulcer debridement. After surgery on April 12, 2016, oxycodone slow release was restarted at 60 mg Q12h with interim doses of 10 mg immediate release Q4h. Concomitant medication included pregabalin 200 mg × 2 p.o. and diazepam 6 mg p.o. Liver and kidney function and other laboratory data were within normal ranges. On April 20, a yeast urinary tract infection was diagnosed. An 800-mg intravenous dose of fluconazole was administered on April 22 at 6 p.m. followed by 400 mg daily. The first 400mg dose was infused on April 23 at 8 a.m. At 10 p.m., the patient’s respiration rate was 8 breaths/min. A 0.4-mg dose of IV naloxone was given and oxycodone was withdrawn. Soon thereafter, his respiration rate increased to 12 breaths/min. He required four additional doses of naloxone on April 24 at 1, 3, and 6 a.m. and 4 p.m. until his respiration rate was maintained over 15 breaths/min. The dosing schedule for the other drugs was unchanged. Because we suspected multiple drug–drug pharmacokinetic and pharmacodynamic interactions including fluconazole, oxycodone, diazepam, and pregabalin, the patient was genotyped for the most common allelic variants of the CYP2D6 and CYP2C19 genes: *3, *4,*5, duplication and *2,*3,*17, respectively. In addition, polymorphisms of CYP3A4 (*1B, *22) and 3A5 (*3, *6 *7) associated with a low or no activity of these cytochromes were screened. He was heterozygous *4 and *10 for CYP2D6 and classified as an intermediate metabolizer; *1/*17 for CYP2C19 and defined as an extensive metabolizer; and was a CYP3A normal metabolizer. The connection between the patient’s decreased rhythm of breathing and the start of fluconazole therapy was striking. Although a fluconazole-oxycodone interaction has never been studied, a possible increase in the average plasma exposure to oxycodone is supported by experimental considerations with regards to fluconazole inhibiting the in vitro metabolism of oxycodone [8]. Respiratory depression occurred 28 h after the first fluconazole infusion. Such a delay was observed in oxycodone-treated patients when voriconazole was introduced [9]. Our patient received 1200 mg of fluconazole over a 14-h time period. Thus, we hypothesized that this dose administered over a short time could have precipitated the onset of this adverse event by inhibiting the CYP3A4 pathway, especially because the shift of oxycodone metabolism to CYP2D6 was unlikely because the patient was an * Bruno Charpiat bruno.charpiat@chu-lyon.fr

Use of Systemic Vasodilators for the Management of Doppler Ultrasound Arterial Abnormalities After Orthotopic Liver Transplantation.

Background Doppler ultrasound (DUS) arterial abnormalities (DAA) after orthotopic liver transplantation (OLT) often represent a sign of hepatic artery (HA) complication (HAC). The standard management of DAA involves computed tomographic angiography (CTA) followed by invasive vascular intervention (IVI) or observation. We evaluated the contribution of systemic vasodilators (SVD) to the management of DAA after OLT. Methods Between 2005 and 2015, 91 of 514 OLT recipients developed DAA (defined by HA resistive index [HARI] <0.5) and received oral SVDs. Doppler ultrasound was performed 2 days later, and patients were assigned to 3 groups accordingly: the normalization group (HARI >0.5), improvement group (HARI increase of >0.1 but value <0.5), or nonresponse group. We analyzed the contribution of this strategy to predict clinically significant HAC, defined as thrombosis or HAC requiring IVI. Results A clinically significant HAC (4 thromboses, 35 HACs requiring IVI) was found in 2.9% (n = 1/34), 32.1% (n = 9/28), and 100% (n = 29/29) of patients in the normalization, improvement, and nonresponse groups, respectively (P < 0.001). On multivariate analysis, absence of HARI normalization after SVD and time to DAA longer than 30 days were associated with clinically significant HAC. Specificity and accuracy of DUS after SVD increased from 88.1% to 95.1% and from 88.9% to 95.1% (P < 0.001), without altering its sensitivity (97.7% vs 95.5%, P = 1.000). Conclusions The use of SVD improves the diagnostic performance of DUS for clinically significant HAC after OLT. It allows identifying patients at low risk for HAC, for whom CTA could be avoided, and helps choosing between observation and IVI in patients with inconclusive CTA.

Hepatic venous pressure gradient after portal vein embolization: An accurate predictor of future liver remnant hypertrophy

Background: The impact of portal hemodynamic variations after portal vein embolization on liver regeneration remains unknown. We studied the correlation between the parameters of hepatic venous pressure measured before and after portal vein embolization and future hypertrophy of the liver remnant after portal vein embolization. Methods: Between 2014 and 2017, we reviewed patients who were eligible for major hepatectomy and who had portal vein embolization. Patients had undergone simultaneous measurement of portal venous pressure and hepatic venous pressure gradient before and after portal vein embolization by direct puncture of portal vein and inferior vena cava. We assessed these parameters to predict future liver remnant hypertrophy. Results: Twenty‐six patients were included. After portal vein embolization, median portal venous pressure (range) increased from 15 (9–24) to 19 (10–27) mm Hg and hepatic venous pressure gradient increased from 5 (0–12) to 8 (0–14) mm Hg. Median future liver remnant volume (range) was 513 (299–933) mL before portal vein embolization versus 724 (499–1279) mL 3 weeks after portal vein embolization, representing a 35% (7.4–83.6) median hypertrophy. Post–portal vein embolization hepatic venous pressure gradient was the most accurate parameter to predict failure of future liver remnant to reach a 30% hypertrophy (c‐statistic: 0.882 [95% CI: 0.727–1.000], P < 0.001). A cut‐off value of post–portal vein embolization hepatic venous pressure gradient of 8 mm Hg showed a sensitivity of 91% (95% CI: 57%–99%), specificity of 80% (95% CI: 52%–96%), positive predictive value of 77% (95% CI: 46%–95%) and negative predictive value of 92.3% (95% CI: 64.0%–99.8%). On multivariate analysis, post–portal vein embolization hepatic venous pressure gradient and previous chemotherapy were identified as predictors of impaired future liver remnant hypertrophy. Conclusion: Post–portal vein embolization hepatic venous pressure gradient is a simple and reproducible tool which accurately predicts future liver remnant hypertrophy after portal vein embolization and allows early detection of patients who may benefit from more aggressive procedures inducing future liver remnant hypertrophy. (Surgery 2018;143:1–2.)

Biliary Duct-to-Duct Reconstruction with a Tunneled Retroperitoneal T-Tube During Liver Transplantation: a Novel Approach to Decrease Biliary Leaks After T-Tube Removal

The benefit of placing a T-tube for duct-to-duct biliary reconstruction during orthotopic liver transplantation (OLT) remains controversial because it could be associated with specific complications, especially at the time of T-tube removal. While the utility of T-tube during OLT represents an eternal debate, only a few technical refinements of T-tube placement have been described since the report of the original technique by Starzl and colleagues. Herein, we present a novel technique of T-tube placement for duct-to-duct biliary reconstruction during OLT, using a tunneled retroperitoneal route. On the basis of our experience of 305 patients who benefitted from the reported technique, the placement of a tunneled retroperitoneal biliary T-tube appears to be safe and results in a low rate of biliary complications, especially at the time of T-tube removal.

Incidence and Risk Factors of Coagulation Profile Derangement After Liver Surgery: Implications for the Use of Epidural Analgesia—a Retrospective Cohort Study

BACKGROUND: Hepatic surgery is a major abdominal surgery. Epidural analgesia may decrease the incidence of postoperative morbidities. Hemostatic disorders frequently occur after hepatic resection. Insertion or withdrawal (whether accidental or not) of an epidural catheter during coagulopathic state may cause an epidural hematoma. The aim of the study is to determine the incidence of coagulopathy after hepatectomy, interfering with epidural catheter removal, and to identify the risk factors related to coagulopathy. METHODS: We performed a retrospective review of a prospective, multicenter, observational database including patients over 18 years old with a history of liver resection. Main collected data were the following: age, preexisting cirrhosis, Child-Pugh class, preoperative and postoperative coagulation profiles, extent of liver resection, blood loss, blood products transfused during surgery. International normalized ratio (INR) ≥1.5 and/or platelet count <80,000/mm3 defined coagulopathy according to the neuraxial anesthesia guidelines. A logistic regression analysis was performed to assess the association between selected factors and a coagulopathic state after hepatic resection. RESULTS: One thousand three hundred seventy-one patients were assessed. Seven hundred fifty-nine patients had data available about postoperative coagulopathy, which was observed in 53.5% [95% confidence interval, 50.0–57.1]. Maximum derangement in INR occurred on the first postoperative day, and platelet count reached a trough peak on postoperative days 2 and 3. In the multivariable analysis, preexisting hepatic cirrhosis (odds ratio [OR] = 2.49 [1.38–4.51]; P = .003), preoperative INR ≥1.3 (OR = 2.39 [1.10–5.17]; P = .027), preoperative platelet count <150 G/L (OR = 3.03 [1.77–5.20]; P = .004), major hepatectomy (OR = 2.96 [2.07–4.23]; P < .001), and estimated intraoperative blood loss ≥1000 mL (OR = 1.85 [1.08–3.18]; P = .025) were associated with postoperative coagulopathy. CONCLUSIONS: Coagulopathy is frequent (53.5% [95% confidence interval, 50.0–57.1]) after liver resection. Epidural analgesia seems safe in patients undergoing minor hepatic resection without preexisting hepatic cirrhosis, showing a normal preoperative INR and platelet count.