Hepatic Surgery in Patients with Cirrhosis: Mitigating Risk



Fig. 12.1
Hepatocellular carcinoma treatment algorithm. HCC hepatocellular carcinoma, CTP Child-Turcotte-Pugh, TACE transarterial chemoembolization, PH portal hypertension, Bili bilirubin; Orange outline indicates palliative treatments only





Colorectal Liver Metastases


While older guidelines suggested that resection for CRLM should be limited to those with four or fewer lesions, and unilobar disease, surgeons are now able to safely pursue resection in both lobes in patients with more substantial numbers of lesions, provided that adequate inflow, outflow, and liver remnant volume is preserved. One study of 484 patients demonstrated that although large numbers of metastases portend poor outcomes, those with eight or more lesions still demonstrated 24% 5-year survivals [19]. Several large-scale studies have looked at ablation versus resection as first-line and second-line therapies for CRLM and have concluded that recurrence rates are unacceptably higher in the ablation group [2022]. Thus, currently, even small singular resectable CRLM are resected preferentially even in cirrhotics. Of course, those with poorly compensated Child’s B cirrhosis, or Child’s C patients are not resection candidates. In these patients, percutaneous ablation is used for local control.


Anatomical Versus Parenchymal Sparing Resections


For any of the indications above, partial liver resections in Child’s A cirrhotics (or with MELD <9) with only mild portal hypertension must leave at least 40% of the previous liver volume behind in order to ensure adequate postoperative liver function [23]. Necessary functional liver remnant volume numbers are less clearly defined, but decidedly larger for those patients with more severe liver disease. Thus, parenchymal sparing techniques are preferred to preserve liver function for those with chronic liver disease (CLD) and portal hypertension. This helps enhance postoperative liver function and preserves parenchyma in case additional procedures are needed in the future. When it comes to HCC, several large-scale studies have demonstrated that survival following liver resection depends on the severity of cirrhosis and tumor features rather than anatomical versus nonanatomical resection [24, 25].



Risk Stratification


Several scores have been developed to grade medical well-being and mortality risks in cirrhotic patients. The Child-Turcotte-Pugh (CTP) classification assigns points to five factors associated with cirrhosis (encephalopathy, ascites, bilirubin, albumin, and international normalized ratio) and assigns a class A, B, or C based on point range. The Model for End-Stage Liver Disease (MELD) uses bilirubin, INR, and creatinine to calculate disease severity. These have been validated in perioperative settings as well to correlate certain scores with perioperative mortality rates [26]. MELD ≥9 portends less favorable operative mortality [27]. The Mayo Clinic has evaluated the contribution of additional features including American Society of Anesthesiologists (ASA) score, etiology of hepatitis (viral vs. alcohol or cholestatic), and patient age in order to calculate scores for individual patients and determine perioperative mortality risk [26]. While these scores fail to take into account major risk factors like portal hypertension, they are very helpful in having honest discussions with patients and their families.

Additional features of clinically significant portal hypertension can be taken into account if the risk scores are falsely low. Surrogate markers for clinically significant portal hypertension include but are not limited to thrombocytopenia (platelets < 100/nL), splenomegaly, endoscopic visualization or cross-sectional imaging showing varices or excessive venous collaterals, clinical signs of collateralization such as caput medusa, and substantial hemorrhoids. That being said, portal hypertension alone should not be considered a contraindication for hepatic resection in cirrhotics [28]. But it should be used as a warning sign that perhaps extent of liver disease has been underestimated. Case-matched controlled studies have demonstrated that MELD score and extent of hepatectomy are the best predictors of perioperative mortality, regardless of the presence or absence of portal hypertension [28].


Risk Modification


Etiology of cirrhosis in patients anticipating liver resection determines preoperative risk modification priorities. In patients undergoing resection for CCA or CRLM, cholestatic or intrinsic hepatic dysfunction can occur as a result of mechanical biliary obstruction or as a result of chemotoxicity from a variety of regimens. Adequate wait times from chemotherapy until surgery can allay the severity of hepatic toxicity. Thus, if there is evidence of chemotherapy-associated steatohepatitis, the authors prefer to wait 6–8 weeks until operative intervention rather than standard 3–4 weeks from last dose of chemotherapy. These recommendations are of course dependent upon patient status and regimens used. For those patients with viral hepatitis-induced cirrhosis, viral load can be associated with recurrence of HCC. Thus, the authors recommend close coordination with a skilled hepatologist, and treatment of Hepatitis C with new curative agents or with older supportive care therapies is necessary to decrease viral load prior to operative intervention.

Operative planning always involves consideration of the extent of resection. Patients with insufficient liver remnant volume to sustain normal liver function are at high risk for postoperative liver failure. In patients with normal liver, a future liver remnant can be 20–25% of preresection liver volume. However, in patients with hepatic failure or cirrhosis, that number is closer to 30–40%. Some authors have set a numerical value of 250 mL/m2 as the minimum liver remnant volume for patients with chronic hepatitis and cirrhosis [29].

Postoperative liver failure (POLF) is perhaps the most significant potential postoperative complication. Risk factors for postoperative liver failure are listed in Table 12.1 [30]. Liver regeneration after major hepatic resection is blunted in patients with cirrhosis or chronic hepatitis [29]. Careful preoperative planning, use of parenchymal sparing techniques, and ablative procedures when resection is impossible can substantially decrease the risk of POLF. Postoperative vigilance in preventing and treating infection and thromboembolic complications can also help avoid POLF. Nutritional status is a modifiable risk factor for poor postoperative outcomes. A large majority of patients with advanced liver disease are malnourished, secondary to defects in protein metabolism [31]. A growing body of literature supports perioperative care pathways that include preoperative nutritional supplementation, with emphasis on protein intake in order to help optimize this risk factor.


Table 12.1
Risk factors and modifications for postoperative liver failure













Nonmodifiable risk factors

Modifiable risk factors

Age > 70

Male

Fibrosis

Preoperative chemotherapy

Steatosis

Diabetes

Portal hypertension

Hepatitis

Excessive intraoperative blood loss

Need for blood transfusion

Prolonged ischemia time

Prolonged operative time

Preoperative hypoalbuminemia

Cholestasis

Coagulopathy can occur via many different mechanisms in liver failure. All coagulation factors with the exception of von Willebrand factor are produced in the liver. Thus, derangement in hepatic function can directly and indirectly effect coagulopathy. Cholestasis and malnutrition can result in vitamin K malabsorption. Portal hypertension resulting in hypersplenism can facilitate platelet trapping and peripheral thrombocytopenia. Thus, supplementation with vitamin K, fresh frozen plasma, and platelets can be helpful in the emergent preoperative setting or in patients with postoperative coagulopathy. However, if an elective resection patient is requiring coagulation factor supplementation, the operation is very likely to cause the patient harm and should therefore be considered very carefully. It is important to note that coagulopathy secondary to chronic liver disease does not protect against venous thromboembolism (VTE) in hospitalized patients with chronic liver disease [32]. Thus, standard perioperative VTE prophylaxis should still be considered.


Surgical Technique


When possible, and particularly in patients with no prior abdominal operations and peripheral liver lesions, minimally invasive surgical techniques should be employed. Use of laparoscopic techniques for resection or ablation has become the standard of care for small (<5 cm) peripheral lesions [33]. Several groups have proven that minimally invasive surgery is safe in patients with mild liver disease [34].

Anticipating and preparing for problems that can arise when operating on cirrhotics is half the battle of a successful resection. Close communication with the anesthesia team in the perioperative setting is critical. We do not recommend use of a central line, but do require two large bore IVs as well as an arterial line in cases where high blood loss is possible. It is critical that central venous pressure (CVP) be maintained at or below 5 mmHg until parenchymal transection is complete [3537]. Restrictive transfusion policies can also be of benefit, and these along with blood product resuscitation protocols should be established with anesthesia colleagues in advance of the surgery [38]. Some of the more challenging aspects of surgery in cirrhotics and ways to combat them are listed in Table 12.2. Generally, surgeons should be prepared for robust collateral veins in unexpected locations with the propensity toward vigorous bleeding secondary to increased portal pressure. The more prominent collaterals are often visible on preoperative imaging. However, meticulous dissection techniques should be employed at all points of dissection in order to avoid vascular injury.


Table 12.2
Technical modifications for surgery in cirrhotics



















Problem

Solution

Bleeding

Medical optimization

Meticulous dissection

Adequate preoperative imaging

Expect collaterals––preop embolization prn

Difficult mobilization

Hand assist

Positioning––use suspensory ligaments and gravity

Postoperative ascites

Preoperative optimization

Postoperative dieresis

Adequate liver remnant

Regarding abdominal entry, standard port placements for left and right robotic and laparoscopic hepatectomies have been described at length [39, 40]. When comparing open with laparoscopic resections in a case-matched study of patients with CLD, one group found that major complications were observable but not significantly more common in the open group, while there was no difference in margin status or long-term survival [34]. For open procedures, the authors generally prefer a so-called inverse L-shaped incision with the option to carry the incision to the left and make a so-called mercedes incision if needed. In cirrhotic patients, great care must be taken to ligate or preserve if possible any recanalized umbilical veins.

In terms of parenchymal transection, some authors advocate use of ultrasonic dissector combined with bipolar electrocautery in cirrhotics [41]. While others note that a less tedious method involves precoagulation and a combination of vascular sealing devices and staplers for larger vessels [42]. In practice, laparoscopically, the authors utilize cautery to enter Glisson’s capsule, and then the laparoscopic LigaSure is used for crush-clamp, sealing, and cutting (Covidien, Minneapolis, MN), with intermittent use of staplers as needed for larger vessels for the vast majority of patients. If a liver is particularly fibrotic, crush-clamp techniques are not as useful and can result in excessive trauma. In these instances, a combination of blunt and thermal dissection techniques is utilized, and a harmonic endoshear device is utilized. The laparoscopic argon beam coagulator is used to obtain hemostasis and biliary stasis from the cut liver’s edge. Of note, both laparoscopically and robotically, Weck clips are utilized rather than metal ones in order to facilitate continued use of vessel-sealing devices. Robotically, the authors utilize similar technique, except that we are able to employ the robotic vessel sealer and stapler tools for added flexibility. For open hepatic resections, the authors employ a combination of crush-clamp and impact LigaSure (Covidien, Minneapolis, MN) for parenchymal transection. Larger vessels are divided with a stapler. When suturing for bleeding, the authors encourage use of silk material, as it does not melt when in contact with thermal cautery or other thermal vascular control agents.

With regard to hilar vascular occlusion (aka the Pringle maneuver), some authors recommend little to no vascular clamping in cirrhotics. Most acknowledge that some cases mandate use of vascular occlusion but that the time should be limited in patients with cirrhosis again to optimize postoperative liver function. Use of intermittent versus continuous clamping has been analyzed by many different authors who have reached disparate conclusions. However, the general consensus is that in terms of morbidity and mortality, there is no difference between the two techniques [37, 43]. That being said, few studies have looked at this issue specifically in cirrhotics.

The authors do not use drains following liver resection as they have been shown in multiple studies to result in increased infectious complications [44]. This remains true in resections of cirrhotics [45]. Should an infected biloma develop postoperatively, our preference is to perform interventional radiology-guided drainage. There is minimal controversy in the literature regarding drain placement following minor resections of noncirrhotic livers; however, there exists some disagreement in the data when cirrhotics with demonstrable preoperative portal hypertension [46]. The authors reserve drain placement for extremely rare occasions such as hepaticojejunostomy creation to tenuous secondary or tertiary radicles.


Outcomes


Increased rates of postoperative morbidity following liver resections in cirrhotics are typically attributed more to poor patient protoplasm than to surgical technique. Cirrhotics have more comorbidities and are at higher risk for POLF and thereby postoperative mortality, more especially following major hepatectomy [42, 47]. POLF is an insidious problem that can result in a frustrating and terrifying march toward postoperative mortality. Because POLF following resection for malignancy can inherently not be remedied with liver transplantation, identification of patients at risk for such a course, and then making every effort to maintain perfusion and biliary continuity in the liver remnant, is critical. Several systems have been developed to identify patients who are on a course toward POLF, the most prominent of which are the so-called 50–50 criteria and the International Study Group of Liver Surgery (ISGLS) criteria [30, 47]. The 50–50 criteria state that patients with bilirubin >50 μmol/L and prothrombin time <50% (INR > 1.7) on postoperative day (POD) 5 have a mortality risk exceeding 50% after hepatectomy [47]. Subsequent prospective validation of the 50–50 criteria has verified this and adds that patients meeting 50–50 criteria on POD 3 are also at high risk for mortality [48]. The ISGLS also utilizes liver function derangements on POD 5 as a guide and further classifies POLF into grades A, B, and C. Risk factors for postoperative mortality taking all comers include meeting the 50–50 criteria, age over 65 years, and presence of severe hepatic fibrosis [47]. Table 12.3 summarizes the two predominant POLF systems.
Jun 27, 2017 | Posted by in GASTROENTEROLOGY | Comments Off on Hepatic Surgery in Patients with Cirrhosis: Mitigating Risk

Full access? Get Clinical Tree

Get Clinical Tree app for offline access