Flu-Like Symptoms Following Radiofrequency Liver Transection: A New Variety of the Post-Radiofrequency Syndrome

Transcription

1 Journal of Investigative Surgery, Early Online, 1 7, 2013 Copyright C 2013 Informa Healthcare USA, Inc. ISSN: print / online DOI: / ARTICLE Flu-Like Symptoms Following Radiofrequency Liver Transection: A New Variety of the Post-Radiofrequency Syndrome Pablo Parra-Membrives, PhD, 1,2 Darío Martínez-Baena, MD, 2 Jose Manuel Lorente-Herce, MD 2 J Invest Surg 1 Department of Surgery, University of Seville, Sevilla, Spain, 2 Hepato-bilio-pancreatic Surgery Unit, General and Digestive Surgery Department, Valme University Hospital, Sevilla, Spain ABSTRACT Background/Aims: The aim of our study was to determine whether post-radiofrequency syndrome may also develop following hepatectomy using saline-cooled radiofrequency coagulation. Methods: We retrospectively reviewed 95 consecutive patients who underwent 110 liver resections between May 2000 and September We stated that 80.9% of the resections were carried out employing the saline-cooled radiofrequency device. All medical records were searched for the occurrence of flu-like symptoms, without evidence of sepsis or infection, in the first two postoperative weeks. Results: Eleven patients (11.5%) developed flu-like symptoms after hepatectomy without evidence of sepsis or infection. All their hepatectomies were performed employing the saline-cooled radiofrequency probe (p =.089), and all cases but one appeared following colorectal liver metastases surgery (p =.042). Eight of them were readmitted to the hospital because of their symptoms. In all 11 cases, a fluid collection was present, 8 of them with gas presence. Nine patients underwent a percutaneous drainage whose cultures were negative. Ten patients recovered without treatment or with the intake of nonsteroidal anti-inflammatory drugs within 1 week, but one patient developed a secondary infection with gram-positive bacteria after percutaneous drainages that prolonged his hospital stay. Conclusion: Liver splitting using saline-cooled radiofrequency coagulation may cause postoperative symptoms that may mimic surgical site infection. Surgeons employing this device should keep this in mind to avoid potentially unwarranted treatments that may be unnecessary, expensive, and even harmful. Keywords: liver surgery; radiofrequency; morbidity; coagulation devices; hepatectomy; flu-like symptoms INTRODUCTION In spite of technical improvements, hepatic surgery is still considered a major procedure with high morbidity rates. Intraoperative blood loss and postoperative complications negatively affect long-term survival [1 3]. Therefore, there have been many efforts to develop an instrument for safe liver transection without bleeding in the last decades. Many devices have been successfully employed in liver surgery, each of them with advantages and disadvantages but with little evidence about the superiority of any of them [4, 5]. In recent years, the saline-cooled radiofrequency probe has gained favor among surgeons performing hepatic resections [6 10]. The device coagulates liver parenchyma allowing for near bloodless liver transection. The probe works by conducting radiofrequency energy from the generator to the electrode tip where continuous low-volume saline irrigation cools the contact surface being cut, keeping the temperature between 100 C and 105 C. This way, the liver tissue is boiled and char formation is avoided. In addition, the thermal energy denatures collagen proteins in the vessel wall and allows blood vessel sealing. However, radiofrequency energy may cause side effects. An inflammatory response causing flu-like symptoms mimicking infection has been described following percutaneous radiofrequency solid tumor ablation, constituting the so-called post-radiofrequency syndrome [11, 12]. To our knowledge, the syndrome has not been associated with saline-cooled radiofrequencyassisted liver transection so far. The purpose of our Received 4 April 2013; accepted 15 July Address correspondence to Dr. Pablo Parra Membrives, C/Rubi 35, Mairena del Aljarafe, Sevilla, Spain. pabloparra@aecirujanos.es 1

2 2 P. Parra-Membrives et al. study was to determine whether post-radiofrequency syndrome is also present after liver surgery employing the saline-cooled radiofrequency device and if radiofrequency associated symptoms may be confused with infectious complications. Patients MATERIAL AND METHODS We retrospectively reviewed 95 consecutive patients (51 men, 44 women; mean age 60.54; range 25 84) who underwent 110 liver resections between May 2000 and September Eight patients received a second hepatectomy because of colorectal liver metastasis recurrence. One of them had a third and a forth recurrence and underwent new liver metastasis resections. Five patients underwent a two-stage hepatectomy. Diagnosis leading to liver resection is listed in Table 1. The average number of resected liver segments was 2.7 (range 1 7). According to the American Society of Anesthesiologists (ASA) score, patients were classified as ASA I in 10.3% of cases, ASA II in 52.9%, ASA III in 33.3%, and ASA IV in 3.4%. Surgical Technique and Perioperative Care All patients received general anesthesia. Patients were monitored with continuous central venous pressure (CVP) measurement, with a low target CVP (0 5 mmhg) during liver transection to minimize blood loss. An epidural catheter was also placed for postoperative analgesia. Antibiotic prophylaxis with 2,200 m of amoxicillin clavulanic acid was administered intravenously on induction of general anesthesia. A surgi- TABLE 1 Diagnosis leading to liver resection Diagnosis Patients Resections Colorectal metastases a Noncolorectal metastases Hepatocellular carcinoma 7 8 Cholangiocarcinoma 7 7 Hemangioma 3 3 Adenoma 3 3 Focal nodular hyperplasia 3 3 Polyciystic liver disease 1 1 Hamartoma 1 1 Chronic intrahepatic cholangitis 1 1 Hydatid cyst 1 1 Xanthogranulomatous cholecytitis 1 1 Total a Seven patients received a second hepatectomy. One of them underwent a third and a forth hepatectomy. Five patients were operated in a two-stage approach and therefore received two hepatectomies. cal technique was conducted as a standard procedure in all cases. A laparotomic approach was used for all patients but one. Our preferred access was a right subcostal J-shaped incision. All patients underwent an intraoperative assessment of their liver disease by ultrasound at the beginning of the surgery, before the liver had been mobilized, to reduce sonographic disturbances. The extend of the resection was then decided. Optimal sparing of parenchyma and negative margins in malignant disease were always the aim during resection planning. Before parenchymal dissection, a tape was passed around the hepatoduodenal ligament for inflow occlusion (Pringle maneuver), though it was used only if there was significant bleeding. If a major liver resection was planned (left/right hepatectomies, extended hepatectomies, or seccionectomies), then inflow vessels were ligated before liver transection was carried out. Our preferred approach was the extra-glissonian approach. Briefly, a TA-30 stapler or an endo-gia stapler, white type, is introduced to transect en-bloc the inflow portal triad. When hilar dissection was necessary, the portal vein, hepatic artery and biliary tract were isolated in the pedicle by opening the Glisson s sheath, and the vascular and biliary structures were transected individually. Minor resections were directly transected without pedicle dissection. Transection was initiated without pedicle clamping. If parenchymal bleeding occurred during liver splitting, 15 min of intermittent hepatic inflow occlusion with a 5-min rest period was applied. Transection was carried out employing a saline-cooled high-frequency monopolar device in combination with the cavitron ultrasonic surgical aspirator (CUSA) in 89 hepatectomies. The saline-cooled probe was used alone in a further case. The remaining liver transections were performed with the aid of a harmonic scalpel in eight cases, by thermocautery in six hepatectomies, using the CUSA alone in five liver splittings, and employing only kellyclasia in three further cases. When the saline-cooled probe was employed together with the CUSA, first, a zone of coagulative necrosis was created with the saline-cooled probe before transecting the liver. After precoagulation, liver splitting and vessel skeletonization were performed with the CUSA device. Coagulation of blood vessels and bile ducts of up to 3 mm in diameter was completed with the saline-cooled radiofrequency probe. Larger intraparenchymal vessels were ligated and cut between interrupted ligatures and metal clips. After transection, the liver surface was coagulated again with saline-cooled radiofrequency until complete hemostasis was achieved. A drainage tube was left in place in 80 procedures. Abdominal drains were removed between 1 and 3 days after surgery depending on the volume and content. Initial postoperative recovery took place in the intensive care unit. In the first postoperative days, analgesia was given through the epidural catheter via a patient-controlled analgesia delivery device. After the catheter was removed, Journal of Investigative Surgery

3 analgesia was given through intravenous administration or oral intake of metamizol or acetaminophen. Oral food intake was allowed in the first 24 hr and patients were discharged from hospital after postoperative complications were excluded, mainly in the first postoperative week. Definitions and Database Review Protocol The database was searched for any complication during hospital stay or any readmittance to hospital during the study period. We analyzed individually all the hospital medical records of the patients who developed complications, with special attention to the presence of flu-like symptoms throughout the in-hospital stay period or during the follow-up in the first three postoperative months. Flu-like symptoms were defined as the transient simultaneous presence of fever (body temperature elevation over 38 C), chills, malaise, myalgia, and nausea or vomiting. Not all symptoms had to be present to be recorded as a possible inflammatory response to radiofrequency. Among these patients, any evidence of sepsis had to be excluded carefully. Pneumonia or respiratory tract infections, urinary tract infections, or wound- and catheter-related infections were excluded by chest X-ray tests or computed tomography (CT) scans, urine sediments and cultures, blood and catheter cultures, and daily wound examinations. Postoperative abdominal CT images were also examined in these patients. Imaging studies had to exclude other surgical site infections. The presence of intra-abdominal fluid collections and the existence of gas associated with them were also recorded. Presence of infection in these collections was examined by percutaneous drainage and bacterial culture of the evacuated fluid. Only patients with transient flu-like symptoms without evidence of infection and percutaneous drained intra-abdominal collections with negative cultures were stated as possibly affected by the postradiofrequency syndrome. Need for hospital readmittance and reoperation due to suspicion of infection was also recorded. Statistical Analysis The SPSS for Mac v20.0 statistical analysis software program was used to carry out chi-square analyses on nominal categorical data. Statistically significant differences were defined as p <.05. RESULTS Eleven patients (11.5% of all patients, six men and five women; mean age 43 years, range years) developed flu-like symptoms after hepatectomy during the first two postoperative weeks. All but one of them had C 2013 Informa Healthcare USA, Inc. Post-Hepatectomy Radiofrequency Syndrome 3 been operated on because of colorectal liver metastases (incidence of 14.5% following hepatectomies because of colorectal metastases vs. 2.4% after liver resections subsequent to other diagnoses, p =.042). All of them had liver transections that were performed with the aid of the saline-cooled radiofrequency probe (12.4% incidence vs. 0% for other liver splitting techniques, p =.082). Symptoms included fever, malaise, and myalgia in nine patients. The remaining two patients experienced malaise, myalgia, and abdominal pain without fever. Symptoms appeared after hospital discharge in all patients. Eight of them were readmitted to the hospital and another three patient were followed up on an outpatient basis. An ultrasonographic study was performed on one of the patients to discard the presence of postoperative abdominal collections. The remaining 10 patients underwent a CT scan. In all cases, a fluid collection in contact with the liver transection surface was revealed. In eight cases, gas presence was noted within the fluid collection (see Figure 1). Nine of eleven patients underwent a CT-guided percutaneous drainage of the suspected abscess obtaining only serous fluid that was negative for bacterial culture. Ten patients recovered without treatment or with the intake of nonsteroidal anti-inflammatory drugs within 1 week of symptom onset, but one patient developed a secondary infection of gram-positive bacteria after percutaneous drainages that prolonged his hospital stay. Complete patient features are listed in Table 2. DISCUSSION Hemorrhage during hepatic surgery is a longstanding concern. The liver receives a rich blood supply through the portal vein and the hepatic artery; therefore, most of the blood loss occurs during liver transection. Given that perioperative transfusion is a wellknown risk factor for poor outcome in hepatic surgery [2], several methods and devices have been employed to minimize blood loss. The dissection with an ultrasonic surgical aspirator has become the procedure used routinely in most centers and many other liver mass coagulators have been employed to aid ultrasonic dissection and minimize blood loss. The argon beam, harmonic scalpel, and Ligasure have shown good results in blood sparing during liver splitting [13 15]. In addition to those, recently the radiofrequency energy has been also applied to liver transection. Although the concept of radiofrequency tip cooling to prevent rapid temperature increase, impedance rises, and char formation is not new and has been used for other indications [16], the Tissuelink device that is employed for liver splitting, a monopolar radiofrequency probe cooled by saline fluid, has given rise to a novel concept: precoagulation [17]. The probe delivers high-frequency energy to the saline fluid that becomes heated and produces an extensive coat of coagulation on the liver

4 4 P. Parra-Membrives et al. J Invest Surg FIGURE 1 A CT scan revealing fluid collection with gas in a patient who underwent a liver resection with a saline-cooled radiofrequency-assisted transection. surface. If coagulation is performed before and after transection, especially in combination with a surgical ultrasonic aspirator, the liver splitting may become virtually bloodless, even in the absence of Pringle vascular occlusion. It is clear that employing the salinecooled radiofrequency probe is not essential for liver transection, and in fact, many surgeons use just clamp crushing for liver surgery in view of the fact that it has proven similar results [5, 11, 12, 18, 19]. However, given that bloodless surgery without ischemia is of paramount importance in living donor hepatic transection or for segment-guided liver resections, the radiofrequency devices have become popular among liver surgeons [6 8, 20 22]. Regardless, radiofrequency energy is not absent of adverse effects. Delayed flu-like symptoms have been reported after percutaneous solid tumor ablation with radiofrequency and have been described as the post-radiofrequency syndrome, whose incidence varies from 31% to 36% of all treated patients [11, 12, 23]. This clinical phenomenon is probably related to inflammatory response to tissue necrosis but not to sepsis onset. The syndrome is characterized by low-grade fever, malaise, myalgia, and nausea and/or vomiting that usually appear within the first hr after the procedure and may last 2 weeks, though some patient may develop late-onset fever. In addition, the presence of gas detected in CT scans after percutaneous radiofrequency use is not new. Some authors have even reported transient and benign portal gas after this treatment, without septic significance [24]. In spite of the fact that the saline-cooled radiofrequency device does not have exactly the same design as the percutaneous radiofrequency ablation probes, both apply the same energy and therefore may cause the same side effects for liver tissue. Saline-cooled radiofrequency induces a deep diffuse, more extensive coagulated liver surface, instead of a spherical necrosed area, with an average thickness of at least 1 cm. This thick layer appears as a linear hypodense demarcation at the surgical margin on CT images and is often misinterpreted as tumor recurrence [25] or as a thick inflammatory abscess wall when fluid is also present. In addition, the saline-cooled probe provides a constant temperature of 100 C at the liver surface that avoids charring and allows greater depths of tissue destruction. However, temperatures may be over 100 C in the subsurface causing steam popping [25, 26]. Therefore, during liver transection, tissue boiling and gas formation are constantly present, and this can easily be seen in any postoperative CT examination and confused with bacterial gas production. All these effects may explain how an inflammatory response to radiofrequency and liver changes induced by the saline-cooled probe may mimic sepsis onset. When a patient presents during the postoperative period with fever and chills and a CT scan revealing fluid collections, microbubbles, and a thick inflammatory layer, the surgeon usually decides to insert a percutaneous drain and to start with antibiotics. However, our study reveals how fever and fluid collection with gas does not always mean infection but may lead to unnecessary antibiotic therapy drain insertion or even surgery. In addition, there is also an economic cost associated with misinterpretation of these effects. Most of the affected patients were readmitted to hospital and Journal of Investigative Surgery

5 TABLE 2 Post-radiofrequency syndrome: patients features Flu-like symptoms Negative Yield of Malaise, bacterial symptoms No. of Intra- chills, culture after No. of resected Drains at operative myalgia, CT scan/ Fluid Gas of fluid NSAID Secondary Hospital Case segments surgery transfusion Fever pain US study collection presence collection intake infection re-admittance NA NA b NA c + NA NA b NA NA NA NA NA a + NA NA NA a Note: NA: data not available; NSAID: nonsteroidal anti-inflammatory drugs. a Patients who underwent a two-stage procedure. b Data not available because percutaneous drainage was not performed. c Patient who underwent an ultrasonographic study. 5

6 6 P. Parra-Membrives et al. underwent many unnecessary investigations. In spite of the fact that none of the patients needed new surgery because of suspected infection, there was a case of iatrogenic infection after repeated percutaneous drain insertion. Given that most of our patients were operated on with the aid of the saline-cooled radiofrequency probe, it could be argued that flu-like symptoms may not be due to the radiofrequency device in particular but simply a general effect of surgery. Although not statistically significant (p =.089), it is remarkable that there was not a single case of flu-like symptoms following the remaining 26 hepatectomies (23.63% of all cases) in which liver splitting was carried out without employing any radiofrequency-related device. A little more than half of our patients were operated on because of colorectal liver metastases, most of whom received preoperative chemotherapy as it is the present standard of care [27]. However, all but one of the cases with post-radiofrequency syndrome in our study were treated because of colorectal liver metastases, which was statistically significant. As the number of cases is still small, we have no data to determine if chemotherapy-treated livers are more prone to develop the radiofrequency syndrome. However, it is consistent with the literature that preoperative chemotherapy predisposes to increased operative blood loss [28] and may have therefore required a greater use of the radiofrequency probe and led to a bigger transection scar. This issue will have to be taken into consideration in the future, and at present, the influence of chemotherapy in the syndrome is only speculative. This study is a preliminary report with retrospective data, but we postulate that an inflammatory response in the manner of post-radiofrequency syndrome was responsible for the clinical features of our patients, as any evidence of sepsis, infection, or any other condition causing flu-like symptoms was reasonably excluded. To our knowledge, we are the first to state this variety of the post-radiofrequency syndrome associated with the saline-cooled radiofrequency probe use. Our reported incidence of 11.5% is a little lower than the incidence following percutaneous tumor ablation. However, due to the retrospective nature of the study, screening of flu-like symptoms was more difficult and the real incidence could have been underestimated. Since all the patients had already been discharged when they presented with symptoms, the patients with only mild symptoms may not have searched for medical assistance. In any case, we believe that our reported incidence is not small, and as a consequence of these findings, we have slightly changed our transection technique. Liver splitting is now carried out at our institution with less precoagulation and coagulation of the remaining-liver to minimize the necrotic scar left behind. Postoperative fever following liver surgery is frequently attributed to infection. However, although persistent or late onset of fever should raise high suspicion of concurrent infection and abscess formation should be ruled out, surgeons dealing with such patients should also consider the possibility of the phenomenon described in this article. This should minimize treatments that may be unnecessary, expensive, and even harmful. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article. REFERENCES [1] Ito H, Are C, Gonen M, et al. Effect of postoperative morbidity on long-term survival after hepatic resection for metastatic colorectal cancer. Ann Surg. 2008;247(6): [2] Kooby DA, Stockman J, Ben-Porat L, et al. Influence of transfusions on perioperative and long-term outcome in patients following hepatic resection for colorectal metastases.ann Surg. 2003;237(6): , discussion [3] Yamamoto J, Kosuge T, Takayama T, et al. Perioperative blood transfusion promotes recurrence of hepatocellular carcinoma after hepatectomy. Surgery. 1994;115(3): [4] Lesurtel M, Selzner M, Petrowsky H, et al. How should transection of the liver be performed?: a prospective randomized study in 100 consecutive patients: comparing four different transection strategies. Ann Surg. 2005;242(6): , discussion [5] Gurusamy KS, Pamecha V, Sharma D, et al. Techniques for liver parenchymal transection in liver resection. Cochrane Database Syst Rev. 2009(1):CD [6]ElMoghazyWM,HedayaMS,KaidoT,etal.Twodifferent methods for donor hepatic transection: cavitron ultrasonic surgical aspirator with bipolar cautery versus cavitron ultrasonic surgical aspirator with radiofrequency coagulator-a randomized controlled trial. Liver Transpl. 2009;15(1): [7] Geller DA, Tsung A, Maheshwari V, et al. Hepatic resection in 170 patients using saline-cooled radiofrequency coagulation. HPB (Oxford). 2005;7(3): [8] Hutchins R, Bertucci M. Experience with TissueLink radiofrequency-assisted parenchymal division. Dig Surg. 2007;24(4): [9] Lee KF, Wong J, Ng W, et al. Feasibility of liver resection without the use of the routine Pringle manoeuver: an analysis of 248 consecutive cases. HPB (Oxford). 2009;11(4): [10] Lin E, Gonzalez R, Venkatesh KR, et al. Can current technology be integrated to facilitate laparoscopic living donor hepatectomy? Surg Endosc. 2003;17(5): [11] Wah TM, Arellano RS, Gervais DA, et al. Image-guided percutaneous radiofrequency ablation and incidence of postradiofrequency ablation syndrome: prospective survey. Radiology. 2005;237(3): [12] Carrafiello G, Lagana D, Ianniello A, et al. Postradiofrequency ablation syndrome after percutaneous radiofrequency of abdominal tumours: one centre experience and review of published works. Australas Radiol. 2007;51(6): [13] Nagano Y, Matsuo K, Kunisaki C, et al. Practical usefulness of ultrasonic surgical aspirator with argon beam coagulation for hepatic parenchymal transection. World J Surg. 2005;29(7): Journal of Investigative Surgery

7 Post-Hepatectomy Radiofrequency Syndrome 7 J Invest Surg [14] Ikeda M, Hasegawa K, Sano K, et al. The vessel sealing system (LigaSure) in hepatic resection: a randomized controlled trial. Ann Surg. 2009;250(2): [15] Arru M, Pulitano C, Aldrighetti L, et al. A prospective evaluation of ultrasonic dissector plus harmonic scalpel in liver resection. Am Surg. 2007;73(3): [16] Shake JG, Larson DW, Salerno CT, et al. The role of electrolyte in lesion size using an irrigated radiofrequency electrode.j Invest Surg. 1997;10(6): , discussion [17] Fioole B, van der Bilt JD, Elias SG, et al. Precoagulation minimizes blood loss during standardized hepatic resection in an experimental model. Br J Surg. 2005;92(11): [18] Zhang S, Zheng Y, Wu B, et al. Is the TissueLink dissecting sealer a better liver resection device than clamp-crushing? A meta-analysis and system review. Hepatogastroenterology. 2012;59(120): [19] Rahbari NN, Koch M, Schmidt T, et al. Meta-analysis of the clamp-crushing technique for transection of the parenchyma in elective hepatic resection: back to where we started? Ann Surg Oncol. 2009;16(3): [20] Geller DA, Tsung A, Maheshwari V, et al. Hepatic resection in 170 patients using saline-cooled radiofrequency coagulation. HPB (Oxford). 2005;7(3): [21] Xia F, Wang S, Ma K, et al. The use of saline-linked radiofrequency dissecting sealer for liver transection in patients with cirrhosis. J Surg Res. 2008;149(1): [22] Di Carlo I, Pulvirenti E, Toro A. Use of dissecting sealer may affect the early outcome in patients submitted to hepatic resection. HPB (Oxford). 2008;10(4): [23] Dodd GD, 3rd, Napier D, Schoolfield JD, et al. Percutaneous radiofrequency ablation of hepatic tumors: postablation syndrome. AJR Am J Roentgenol. 2005;185(1): [24] Oei T, vansonnenberg E, Shankar S, et al. Radiofrequency ablation of liver tumors: a new cause of benign portal venous gas. Radiology. 2005;237(2): [25] McGahan JP, Khatri VP. Imaging findings after liver resection by using radiofrequency parenchymal coagulation devices:initial experiences.radiology. 2008;247(3): [26] Chang I. Finite element analysis of hepatic radiofrequency ablation probes using temperature-dependent electrical conductivity. Biomed Eng Online. 2003;2:12. [27] Nordlinger B, Sorbye H, Glimelius B, et al. Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet. 2008;371(9617): [28] Kneuertz PJ, Maithel SK, Staley CA, et al. Chemotherapyassociated liver injury: impact on surgical management of colorectal cancer liver metastases. Ann Surg Oncol. 2010;18(1): C 2013 Informa Healthcare USA, Inc.