Treating Type 1 Diabetes by Pancreas Transplant Alone: A Cohort Study on Actual Long-term (10 Years) Efficacy and Safety.

BACKGROUND: Physiologically regulated insulin secretion and euglycemia are achievable in type 1 diabetes (T1D) by islet or pancreas transplantation. However, pancreas transplant alone (PTA) remains a debated approach, with uncertainties on its relative benefits and risks. We determined the actual long-term (10 y) efficacy and safety of PTA in carefully characterized T1D subjects. METHODS: This is a single-center, cohort study in 66 consecutive T1D subjects who received a PTA between April 2001 and December 2007, and were then all followed until 10 y since transplant. Main features evaluated were patient survival, pancreas graft function, C-peptide levels, glycemic parameters, and the function of the native kidneys. RESULTS: Ten-year actual patient survival was 92.4%. Optimal (insulin independence) or good (minimal insulin requirement) graft function was observed in 57.4% and 3.2% of patients, respectively. Six (9.0%) patients developed stage 5 or 4 chronic kidney disease. In the remaining individuals bearing a successful PTA, estimated glomerular filtration rate (eGFR) decline per year was -2.29 ± 2.69 mL/min/1.73 m2. Reduction of eGFR at 1 y post-PTA was higher in those with pre-PTA hyperfiltration and higher HbA1c concentrations; eGFR changes afterward significantly correlated with diabetes duration. In recipients with normoglycemia at 10 y, 74% of normoalbuminuric or microalbuminuric subjects pre-PTA remained stable, and 26% progressed toward a worse stage; conversely, in 62.5% of the macroalbuminuric individuals albuminuria severity regressed. CONCLUSIONS: These long-term effects of PTA on patient survival, graft function, and the native kidneys support PTA as a suitable approach to treat diabetes in selected T1D patients.

Joint consensus on abdominal robotic surgery and anesthesia from a task force of the SIAARTI and SIC.

Minimally invasive surgical procedures have revolutionized the world of surgery in the past decades. While laparoscopy, the first minimally invasive surgical technique to be developed, is widely used and has been addressed by several guidelines and recommendations, the implementation of robotic-assisted surgery is still hindered by the lack of consensus documents that support healthcare professionals in the management of this novel surgical procedure. Here we summarize the available evidence and provide expert opinion aimed at improving the implementation and resolution of issues derived from robotic abdominal surgery procedures. A joint task force of Italian surgeons, anesthesiologists and clinical epidemiologists reviewed the available evidence on robotic abdominal surgery. Recommendations were graded according to the strength of evidence. Statements and recommendations are provided for general issues regarding robotic abdominal surgery, operating theatre organization, preoperative patient assessment and preparation, intraoperative management, and postoperative procedures and discharge. The consensus document provides evidence-based recommendations and expert statements aimed at improving the implementation and management of robotic abdominal surgery.

Update on pancreatic transplantation on the management of diabetes.

Pancreas transplantation is the only therapy that can restore insulin independence in beta-cell penic diabetic recipients. Because of the need for life-long immunosuppression and the intial surgical risk associated with the transplant procedure, Pancreas transplantation is a therapeutic option only in selected diabetic patients. Based on renal function, three main populations of diabetic recipients of a pancreas transplant can be identified: uremic patients, posturemic patients (following successful kidney transplantation), and non-uremic patients. Uremic patients are best treated by simultaneous kidney-pancreas transplantation with grafts obtained from the same deceased donor. Posturemic patients can receive a pancreas after kidney transplantation, if the previous renal graft has a good functional reserve. Non-uremic patients can receive a pancreas alone transplant if their diabetes is poorly controlled, despite optimal insulin therapy, suffer from unawareness hypoglycemia events and/or develop progressive chronic complications of diabetes. The results of pancreas transplantation have improved over the years and are currently not inferior to those of renal transplantation in non-diabetic recipients. A functioning pancreatic graft can prolong the life of diabetic recipients, improves their quality of life, and can halt, or reverse, the progression of chronic complications of diabetes. Unfortunately, because of ageing of donor population and lack of timely referral of potential recipients, the annual volume of pancreas transplants is declining. Considering that the results of pancreas transplantation depend on center volume, and that adequate center volume is required also for training of newer generations of transplant physcians and surgeons, centralization of pancreas transplantation activity should be considered.

Influence of videogames and musical instruments on performances at a simulator for robotic surgery.

OBJECTIVE: To assess if exposure to videogames, musical instrument playing, or both influence the psychomotor skills level, assessed by a virtual reality simulator for robot-assisted surgery (RAS). MATERIALS AND METHODS: A cohort of 57 medical students were recruited: playing musical instruments (group 1), videogames (group 2), both (group 3), and no activity (group 4); all students executed four exercises on a virtual simulator for RAS. RESULTS: Subjects from group 3 achieved the best performances on overall score: 527.09 ± 130.54 vs. 493.73 ± 108.88 (group 2), 472.72 ± 85.31 (group 1), and 403.13 ± 99.83 (group 4). Statistically significant differences (p < .05) between group 3 and group 4 were found for overall score (p = .009) and for time of completion (p = .044). As regards experience with the piano, subjects from group 3 outperformed those from group 1 on overall score (496.98 ± 122.71 vs. 470.25 ± 92.31), but without statistically significant difference (p = .646). CONCLUSIONS: The present study suggests that the level of psychomotor skills in subjects exposed to both musical instrument playing and videogames is higher than that in those practicing either one alone. The effect of videogames appears negligible in individuals playing the piano.

Indications, technique, and results of robotic pancreatoduodenectomy.

Robotic assistance improves surgical dexterity in minimally invasive operations, especially when fine dissection and multiple sutures are required. As such, robotic assistance could be rewarding in the setting of robotic pancreatoduodenectomy (RPD). RPD was implemented at a high volume center with preemptive experience in advanced laparoscopy. Indications, surgical technique, and results of RPD are discussed against the background of current literature. RPD was performed in 112 consecutive patients. Conversion to open surgery was required in three patients, despite nine required segmental resection and reconstruction of the superior mesenteric/portal vein. No patient was converted to laparoscopy. A pancreato-jejunostomy was created in 106 patients (94.6 %), using either a duct-to-mucosa (n = 82; 73.2 %) or an invaginating (n = 24; 21.4 %) technique. Pancreato-gastrostomy was performed in one patient, the pancreatic duct was occluded in two patients, and a pancreatico-cutaneous fistula was created in three patients. Mean operative time was 526.3 ± 102.4 in the entire cohort and reduced significantly over the course of time. Experience was also associated with reduced rates of delayed gastric emptying and increased proportion of malignant tumor histology. Ninety day mortality was 3.6 %. Postoperative complications occurred in 83 patients (74.1 %) with a median comprehensive complication index of 20.9 (0-30.8). Clinically relevant pancreatic fistula occurred in 19.6 % of the patients. No grade C pancreatic fistula was noted in the last 72 consecutive patients. RPD is safely feasible in selected patients. Implementation of RPD requires sound experience with open pancreatoduodenectomy and advanced laparoscopic procedures, as well as specific training with the robotic platform.

Laparoscopic robot-assisted resection of tumors located in posterosuperior liver segments.

Laparoscopic resection of liver tumors located in the posterosuperior segments is a challenging operation that could be facilitated by robotic assistance. Laparoscopic resection of 12 tumors located in posterosuperior segments (IVa: 1; VII: 5; VIII: 6) was carried out under robotic assistance. All patients had a single tumor nodule. Data were collected prospectively and analyzed retrospectively. Surgery required a mean of 260.4 min (115-430) and was completed laparoscopically in all but one patient, who required conversion to mini-laparotomy because of intolerance of pneumoperitoneum (8.3%). Mean estimated blood loss was 252.7 ml (50-600), making transfusion necessary in 3 patients (25.0%). Post-operative complications occurred in 4 patients (33.3%), being of Clavien-Dindo grade II in 3 patients (25.0%) and Clavien-Dindo grade IV in 1 patient (8.3%). Reoperation was required in 1 patient, who subsequently had a long hospital stay, because of decompensated cirrhosis. Median length of hospital stay was 8.5 days (7-96). No patient was readmitted. Pathology showed hepatocellular carcinoma in 7 patients (58.3%), liver metastasis in 2 patients (16.6%), and hepatic adenoma, focal nodular hyperplasia, and hemangioma in one patient each (8.3%). All patients had a margin negative resection. After a mean follow-up period of 21.4 months (±24.4), no patient with malignant histology developed recurrence. Our initial experience confirms that laparoscopic robot-assisted resection of tumors located in the posterosuperior segments is feasible. Further experience is needed before final conclusions can be drawn and meaningful comparison with other surgical techniques becomes possible.

The learning curve in robotic distal pancreatectomy.

No data are available on the learning curve in robotic distal pancreatectomy (RADP). The learning curve in RADP was assessed in 55 consecutive patients using the cumulative sum method, based on operative time. Data were extracted from a prospectively maintained database and analyzed retrospectively considering all events occurring within 90 days of surgery. No operation was converted to laparoscopic or open surgery and no patient died. Post-operative complications occurred in 34 patients (61.8%), being of Clavien-Dindo grade I-II in 32 patients (58.1%), including pancreatic fistula in 29 patients (52.7%). No grade C pancreatic fistula occurred. Four patients received blood transfusions (7.2%), three were readmitted (5.4%) and one required repeat surgery (1.8%). Based on the reduction of operative times (421.1 ± 20.5 vs 248.9 ± 9.3 min; p < 0.0001), completion of the learning curve was achieved after ten operations. Operative time of the first 10 operations was associated with a positive slope (0.47 + 1.78* case number; R (2) 0.97; p < 0.0001*), while that of the following 45 procedures showed a negative slope (23.52 - 0.39* case number; R (2) 0.97; p < 0.0001*). After completion of the learning curve, more patients had a malignant histology (0 vs 35.6%; p = 0.002), accounting for both higher lymph node yields (11.1 ± 12.2 vs 20.9 ± 18.5) (p = 0.04) and lower rate of spleen preservation (90 vs 55.6%) (p = 0.04). RADP was safely feasible in selected patients and the learning curve was completed after ten operations. Improvement in clinical outcome was not demonstrated, probably because of the limited occurrence of outcome comparators.

Laparoscopic pancreaticoduodenectomy: a systematic literature review.

BACKGROUND: Laparoscopic pancreaticoduodenectomy (LPD) is gaining momentum, but there is still uncertainty regarding its safety, reproducibility, and oncologic appropriateness. This review assesses the current status of LPD. METHODS: Our literature review was conducted in Pubmed. Articles written in English containing five or more LPD were selected. RESULTS: Twenty-five articles matched the review criteria. Out of a total of 746 LPD, 341 were reported between 1997 and 2011 and 405 (54.2 %) between 2012 and June 1, 2013. Pure laparoscopy (PL) was used in 386 patients (51.7 %), robotic assistance (RA) in 234 (31.3 %), laparoscopic assistance (LA) in 121 (16.2 %), and hand assistance in 5 (0.6 %). PL was associated with shorter operative time, reduced blood loss, and lower rate of pancreatic fistula (vs LA and RA). LA was associated with shorter operative time (vs RA), but with higher blood loss and increased incidence of pancreatic fistula (vs PL and RA). Conversion to open surgery was required in 64 LPD (9.1 %). Operative time averaged 464.3 min (338-710) and estimated blood 320.7 mL (74-642). Cumulative morbidity was 41.2 %, and pancreatic fistula was reported in 22.3 % of patients (4.5-52.3 %). Mean length of hospital stay was 13.6 days (7-23), showing geographic variability (21.9 days in Europe, 13.0 days in Asia, and 9.4 days in the US). Operative mortality was 1.9 %, including one intraoperative death. No difference was noted in conversion rate, incidence of pancreatic fistula, morbidity, and mortality when comparing results from larger (≥30 LPD) and smaller (≤29 LPD) series. Pathology demonstrated ductal adenocarcinoma in 30.6 % of the specimens, other malignant tumors in 51.7 %, and benign tumor/disease in 17.5 %. The mean number of lymph nodes examined was 14.4 (7-32), and the rate of microscopically positive tumor margin was 4.4 %. CONCLUSIONS: In selected patients, operated on by expert laparoscopic pancreatic surgeons, LPD is feasible and safe.

Laparoscopic robot-assisted versus open total pancreatectomy: a case-matched study.

BACKGROUND: The enhanced dexterity offered by robotic assistance could be excessive for distal pancreatectomy but not enough to improve the outcome of laparoscopic pancreaticoduodenectomy. Total pancreatectomy retains the challenges of uncinate process dissection and digestive reconstruction, but avoids the risk of pancreatic fistula, and could be a suitable operation to highlight the advantages of robotic assistance in pancreatic resections. METHODS: Eleven laparoscopic robot-assisted total pancreatectomies (LRATP) were compared to 11 case-matched open total pancreatectomies. All operations were performed by one surgeon during the same period of time. Robotic assistance was employed in half of the patients, based on robot availability at the time of surgery. Variables examined included age, sex, American Society of Anesthesiologists score, body mass index, estimated blood loss, need for blood transfusions, operative time, tumor type, tumor size, number of examined lymph nodes, margin status, post-operative complications, 90-day or in-hospital mortality, length of hospital stay, and readmission rate. RESULTS: No LRATP was converted to conventional laparoscopy, hand-assisted laparoscopy or open surgery despite two patients (18.1 %) required vein resection and reconstruction. LRATP was associated with longer mean operative time (600 vs. 469 min; p = 0.014) but decreased mean blood loss (220 vs. 705; p = 0.004) than open surgery. Post-operative complications occurred in similar percentages after LRATP and open surgery. Complications occurring in most patients (5/7) after LRATP were of mild severity (Clavien-Dindo grade I and II). One patient required repeat laparoscopic surgery after LRATP, to drain a fluid collection not amenable to percutaneous catheter drainage. One further patient from the open group required repeat surgery because of bleeding. No patient had margin positive resection, and the mean number of examined lymph nodes was 45 after LRATP and 36 after open surgery. CONCLUSIONS: LRATP is feasible in selected patients, but further experience is needed to draw final conclusions.

High-throughput microRNA (miRNAs) arrays unravel the prognostic role of MiR-211 in pancreatic cancer.

BACKGROUND: Only a subset of radically resected pancreatic ductal adenocarcinoma (PDAC) patients benefit from chemotherapy, and identification of prognostic factors is warranted. Recently miRNAs emerged as diagnostic biomarkers and innovative therapeutic targets, while high-throughput arrays are opening new opportunities to evaluate whether they can predict clinical outcome. The present study evaluated whether comprehensive miRNA expression profiling correlated with overall survival (OS) in resected PDAC patients. METHODOLOGY/PRINCIPAL FINDINGS: High-resolution miRNA profiles were obtained with the Toray's 3D-Gene™-miRNA-chip, detecting more than 1200 human miRNAs. RNA was successfully isolated from paraffin-embedded primary tumors of 19 out of 26 stage-pT3N1 homogeneously treated patients (adjuvant gemcitabine 1000 mg/m(2)/day, days-1/8/15, every 28 days), carefully selected according to their outcome (OS<12 (N = 13) vs. OS>30 months (N = 6), i.e. short/long-OS). Highly stringent statistics included t-test, distance matrix with Spearman-ranked correlation, and iterative approaches. Unsupervised hierarchical analysis revealed that PDACs clustered according to their short/long-OS classification, while the feature selection algorithm RELIEF identified the top 4 discriminating miRNAs between the two groups. These miRNAs target more than 1500 transcripts, including 169 targeted by two or more. MiR-211 emerged as the best discriminating miRNA, with significantly higher expression in long- vs. short-OS patients. The expression of this miRNA was subsequently assessed by quantitative-PCR in an independent cohort of laser-microdissected PDACs from 60 resected patients treated with the same gemcitabine regimen. Patients with low miR-211 expression according to median value had a significantly shorter median OS (14.8, 95%CI = 13.1-16.5, vs. 25.7 months, 95%CI = 16.2-35.1, log-rank-P = 0.004). Multivariate analysis demonstrated that low miR-211 expression was an independent factor of poor prognosis (hazard ratio 2.3, P = 0.03) after adjusting for all the factors influencing outcome. CONCLUSIONS/SIGNIFICANCE: Through comprehensive microarray analysis and PCR validation we identified miR-211 as a prognostic factor in resected PDAC. These results prompt further prospective studies and research on the biological role of miR-211 in PDAC.

Pancreas transplant as treatment to arrest renal function decline in patients with type 1 diabetes and proteinuria.

Recent findings from the Diabetes Control and Complications Trial and Epidemiology of Diabetes Interventions and Complications study showed that long-term improved glycemic control in patients with type 1 diabetes with normal renal function and normoalbuminuria can delay development of impaired renal function by at least 6.5 years, although the reduction in the relative risk of end-stage renal disease (ESRD) was not significant. The unanswered question is: can improvement of glycemic control delay the onset of ESRD in patients with established diabetic nephropathy? In this context, pancreas transplantation (PATx) can be considered the most effective intervention to restore normoglycemia. Can this aggressive/experimental intervention be applied to arrest/retard renal function decline? To answer this question, this review summarizes the relevant findings from observational studies conducted in cohorts of patients, followed up for 4 to 15 years, who underwent PATx. These noncontrolled studies provided positive answers to the earlier question, principally concerning a significant decrease in albumin excretion levels. However, current drugs used to prevent rejection could impair renal function, principally in recipients with low pretransplant estimated glomerular filtration rate (ie, <60 mL/min). Unfortunately, all these studies had shortcuts that qualify interpretation of the findings. First, it is unclear how much initial estimated glomerular filtration rate loss results from nephrotoxic effect of antirejection drugs, and how much results from improved glycemia and its impact on the reduction of hyperfiltration. Second, the study designs did not consider the wide variation in rates of renal function loss observed in patients with established nephropathy (ie, one third are nonprogressors, one third are slow progressors, and one third are rapid progressors). Third, all studies were observational in nature and clinical trials are needed to properly evaluate the effectiveness of normalization of hyperglycemia through PATx on postponing the onset of ESRD in type 1 diabetes.

Transplantation of the pancreas.

Pancreas transplantation consistently induces insulin-independence in beta-cell-penic diabetic patients, but at the cost of major surgery and life-long immunosuppression. One year after grafting, patient survival rate now exceeds 95 % across recipient categories, while insulin independence is maintained in some 85 % of simultaneous pancreas and kidney recipients and in nearly 80 % of solitary pancreas transplant recipients. The half-life of the pancreas graft currently averages 16.7 years, being the longest among extrarenal grafts, and substantially matching the one of renal grafts from deceased donors. The difference between expected (100 %) and actual insulin-independence rate is mostly explained by technical failure in the postoperative phase, and rejection in the long-term period. Death with a functioning graft remains a further major issue, especially in uremic patients who have undergone prolonged periods of dialysis. Refinements in graft preservation, surgical techniques, immunosuppression, and prophylactic treatments are expected to further improve the results of pancreas transplantation.

Central pancreatectomy with inframesocolic pancreatojejunostomy.

PURPOSE: Pancreatic fistula (PF) occurs frequently after central pancreatectomy (CP), but it is not clear from which pancreatic stump it arises and, consequently, which interventions can reduce its incidence and severity. The information could be obtained if the two pancreatic remnants were segregated into different body compartments. METHODS: In eight consecutive patients, the cut end of the distal pancreatic stump after CP was brought in the inframesocolic compartment through a small defect created in the transverse mesocolon. Pancreatojejunostomy was hence constructed in the intraperitoneal compartment, being divided by the retroperitoneal right-sided pancreatic stump by the transverse mesocolon itself. Five patients were operated on open, and three by robot-assisted laparoscopy. PF was defined according to the criteria proposed by the International Study Group on Pancreatic Fistula. RESULTS: PF fistula developed in five out eight patients (three grade A and two grade B). Amylase concentration in the fluid obtained from surgical drains showed that the two pancreatic remnants were actually segregated into different body compartments and that four out of five PF originated from the right remnant. Mean hospital stay was 12.5 days. No patient was readmitted, developed peripancreatic fluid collections, required interventional radiology procedures, or underwent repeat surgery. CONCLUSIONS: In CP, interposing an anatomic barrier, such as the transverse mesocolon, between the two pancreatic remnants is a simple maneuver that, if on one hand, adds little to the complexity of the operation, on the other, provides insights into the origin of PF after CP.

Incidence, diagnosis, and treatment of chylous leakage after laparoscopic live donor nephrectomy.

BACKGROUND: Chylous leakage (CL) is a rare complication of laparoscopic live donor nephrectomy (LLDN). It may lead to malnutrition and immunological deficits because of protein and lymphocyte depletion. METHODS: Data from 208 consecutive LLDN performed at two institutions, between April 2000 and September 2010, were reviewed to identify the anatomical basis behind CL along with its diagnostic and therapeutic options. RESULTS: CL developed in eight donors (3.8%), as determined by high-volume drainage (range 540-800 mL/24 hr) of triglyceride-rich fluid. All donors were managed conservatively. Seven were put on total parenteral nutrition plus octreotide. One received low-fat diet, medium-chain triglyceride supplementation, and octreotide. Chylous fistulas resolved in 5 to 16 days (mean time 12.3 days). Drains were removed before hospital discharge, and no donor was readmitted and/or needed outpatient care. CONCLUSIONS: CL is a potentially insidious and perhaps misdiagnosed complication after LLDN. It occurs in nearly 4% of LLDN and it seems to be uniquely associated to left-sided kidney recovery because of distinctive lymphatics distribution around the periaortic area of dissection. Conservative therapy is effective in most donors and should be initially attempted. Surgical ligatures or fibrin sealants may be indicated in case of refractory CL before the arising of malnutrition and/or relevant immunodeficiency.

Current perspectives on laparoscopic robot-assisted pancreas and pancreas-kidney transplantation.

Pancreas transplant recipients continue to suffer high surgical morbidity. Current robotic technology provides a unique opportunity to test whether laparoscopy can improve the post-operative course of pancreas transplantation (PT). Current knowledge on robotic pancreas and renal transplantation was reviewed to determine feasibility and safety of robotic PT. Information available from literature was included in this review, together with personal experience including three PT, and two renal allotransplants. As of April 2011, the relevant literature provides two case reports on robotic renal transplantation. The author's experience consists of one further renal allotransplantation, two solitary PT, and one simultaneous pancreas-kidney transplantation. Information obtained at international conferences include several other renal allotransplants, but no additional PT. Preliminary data show that PT is feasible laparoscopically under robotic assistance, but raises concerns regarding the effects of increased warm ischemia time on graft viability. Indeed, during construction of vascular anastomoses, graft temperature progressively increases, since maintenance of a stable graft temperature is difficult to achieve laparoscopically. There is no proof that progressive graft warming produces actual damage to transplanted organs, unless exceedingly long. However, this important question is likely to elicit a vibrant discussion in the transplant community.

Renal cell carcinoma with caval involvement: contemporary strategies of surgical treatment.

OBJECTIVES: We retrospectively evaluated the outcome of the surgical treatment of patients with renal cell carcinoma (RCC) and extensive inferior vena cava (IVC) involvement. Our aim was to investigate if a particular surgical technique could reduce morbidity and complications associated with this condition. MATERIALS AND METHODS: From 1996 to 2007, 22 patients with RCC and extensive IVC involvement underwent radical surgical treatment with the intention to avoid, whenever possible, sternotomy and cardiopulmonary bypass. The level of the tumor thrombus was I (<2 cm above the renal vein) in 2 patients, II (below the intrahepatic vena cava) in 9 patients, III (intrahepatic vena cava below the diaphragm) in 7 patients, and IV (atrial) in 4 patients. Extracorporeal vascular bypass was used for 4 patients with level IV and for 2 patients with level III tumor thrombi, with hypothermic circulatory arrest in 2 patients. Extensive liver mobilization techniques were adopted in 16 patients. Overall and cancer-specific survival (CSS) were analyzed based on tumor extent (N0M0, N+M+), pathologic stage (pT3b, pT3c, pT4), thrombus level, and caval wall infiltration. RESULTS: Two patients died within 1 month of surgery and the remaining 20 patients have a mean follow-up of 32.2 months (range 6-90): 8 are alive (overall survival 40%), but 2 with disease (CSS 30%). A total of 10 severe complications developed in 8 patients (36%). Both overall and CSS were significantly associated with tumor stage (Log-rank P = 0.0237 and 0.0465), presence of nodal or systemic metastases (Log-rank P = 0.0835 and 0.0669; Wilcoxon's test P = 0.0407 and 0.0411), and caval wall infiltration (Log-rank P = 0.0200 and 0.0418). CONCLUSIONS: Despite the low overall survival, related to the high percentage of nodal and systemic metastases, aggressive surgical management with resection of synchronous metastatic disease for symptom palliation and cytoreduction, followed by immunotherapy is justified in this setting. A transabdominal approach to RCC and IVC involvement, even in patients with level III thrombus, can provide the surgeon with an exposure similar to thoracoabdominal incisions without the complications associated with thoracotomy.