et al

Document Sample
et al Powered By Docstoc
					    Updates on surgery for
      pancreatic cancer

Paul B. S. Lai
Division of Hepato-biliary and Pancreatic Surgery
Department of Surgery
Chinese University of Hong Kong
      Prognosis of pancreatic cancer

 Typically present late in the course of disease and
  they are usually inoperable
   – locally advance
   – presence of metastasis
 For those patients with resection done
   – high operative mortality
   – high recurrence rate
 Pancreatic cancer
     = Death sentence?
  Resection rates for pancreatic cancer
                            Study            No. of                            Resection rate
Reference         Year      Period          patients                Region          (%)

Bramhall et al.   1995    1977-1987             5672       West Midlands, UK        2·6

NYCRIS            2000    1986-1994             3278       Northern and             4·0
                                                           Yorkshire, UK
Hedberg et al.    1998    1977-1991              575       Mälmo, Sweden            4·2

Sener et al.      1999    1985-1995          100 313       USA                      9·0

   NYCRIS, Northern and Yorkshire Cancer Registry and Information Service.
            Assessment of resectability

 CT scan
  – Remain the gold standard with 80 to 90% accuracy in
    predicting resectability
  – Can be further improved with the use of multi-slice
    scanners with arterial and portal venous phases of
    contrast enhancement
  – Can only detect cancers when they impinge on
    pancreatic duct
  – Small early cancers and small uncinate tumours may be
  – Highly sensitive in picking up small lesions and invasion
    to major vascular structures
  Laparoscopy and laparoscopic USG

 Can detect occult metastatic lesions in the
  liver and peritoneal cavity in 10-35% of
  cases not identified by other imaging
 True value remains controversial
IVC invasion not detected on CT scan


     Surgical resection -
What to do and what not to do?
 Patients with obstructive jaundice

Should we drain the biliary system
   before further intervention?
       Pre-operative biliary drainage

 No clear benefits or harms in terms of surgical
 Memorial Sloan-Kettering reported an increased
  overall morbidity and mortality after pre-operative
 Others found slight increase in wound infection
 Clearly, a lot of units would be forced to stent
  their patients pre-operative for logistic reasons
 Trials are still underway
Pre-operative management
 Imaging
   – CT scan mostly
   – MRCP (not a routine)
   – Cholangiogram (from ERCP or PTBD)
 Drainage of biliary tree pre-op
   – Usually drained due to waiting list consideration
 Prophylactic octreotide
   – Optional
 Bowel preparation
   – Not a routine
 Informed consent
   – Quoting 5% mortality
Union International Contre la Cancer staging
of pancreatic cancer (2002)
            STAGE          T-staging     N-staging   M-staging

              0               Tis               0        0
              IA              T1                0        0
              IB              T2                0        0
              IIA             T3                0        0
              IIB            T1-3           N1           0
              III             T4           Any           0
              IV             Any           Any          M1

Tis, carcinoma in situ;                              N0, no regional lymph
T1, tumour limited to the pancreas < 2 cm;           node metastasis;
T2, tumour limited to the pancreas > 2 cm;           N1, regional lymph node
T3, tumour extending beyond the pancreas             metastasis;
but without involvement of the coeliac axis
or the superior mesenteric artery;
T4, tumour involving coeliac axis or superior        M0, no distant
mesenteric artery;                                   metastasis;
                                                     M1, distant metastasis.
             Operative procedures
 Routine laparoscopy
  – For detection of small peritoneal metastases
 Routine laparoscopic USG assessment
  – To assess for small liver lesions and the status
    of the SMV and PV
 Incision
  – Roof-top
 Determination of operability
  – Mobilization of pancreatic head and
  – Exploration of the lesser sac
  – Cholecystectomy and transection of CHD
Increasing use of
TA 90 to transect
uncinate flush to
SMA by palpation
Assess the
root of
for tumour
Kocher’s manoeuvre
Transection of
CHD above
cystic duct
insertion to
expose the PV

                 Tunneling of the anterior surface of PV
To transect
antrum or D1
      Standard Whipple versus PPPD

 3 prospective randomized trials of Surgery 2004
   – Taiwan (Lin et al 1999; BJS) – 31 patients only
   – Bern (Seiler et al 2004; BJS) – 66 vs. 64
   – Netherlands (Tran et al 2004; Ann Surg) – 83 vs. 87

 Very similar outcomes
 No significant difference in long-term survival
  How to reconstruct pancreatic-enteric
             anastomosis ?

 End-to-side duct-to-mucosa technique with or
  without stent
   – Verona study showed no difference between duct-to-
     mucosa and end-to-side technique

                    Bassi C. et al., Surgery 2003; 134:766-771
• Prospective series of 123 patients
• Cut surface of pancreas evaluated for blood supply
• Those deemed inadequate would have the pancreas cut back
  to improve the blood supply
• Leakage rate = 1.6% (2 in 123 patient)
                      Strasberg et al., J Am Coll Surg 2002;194:746-760.
Anatomy of the pancreatic neck and uncinate process
 Pancreatico-jejunostomy –
duct-to-mucosa & end-to-side
 Pancreatico-jejunostomy –
duct-to-mucosa & end-to-side
  How to reconstruct pancreatic-enteric
             anastomosis ?

 Pancreatojejunostomy (PJ) vs.
  Pancreatogastrostomy (PG)
   – Johns Hopkins prospective randomized study showed
     similar leakage rate (11% vs. 12%) with no peri-operative
   – Another retrospective study of 441 patients from Germany
     showed less leakage and lower mortality associated with
     leakage in the PG group

                  Yeo CJ et al., Ann Surg 1995; 222:580-588.
                  Schlitt HJ et al., BJS 2002; 89:1245-1251.
Radical enough?
 Post-operative management
 ICU not routine
 A short course of antibiotics
   – Cefuroxime, metronidazole
 Octreotide SC
   – Not a routine, sometimes used for insecure PJ
 Drains removed after confirming low
  amylase content
 NG removed after output less than ~200ml a
 Management of complications
 PJ leak
    – Drainage + TPN
    – Relaparotomy in selected cases
 Intraabdominal collections
    – Percutaneous drainage usually adequate
 Wound infection
    – Rather common
    – ? Would protection
    – ? Delayed primary closure
 Delay gastric emptying
    – Functional versus technical
    – Usually settles with conservative treatment
 Bile leak
    – Quite rare, drainage should be adequate
 Post-operative bleeding
    – Rare, but need re-laparotomy if it does occur
    – Association with PJ leakage?
Post-op adjuvant therapy – in PWH

 Not a routine
 No standard protocol
   – RT alone, chemotherapy alone, chemo-RT combined
 Overall survival of patients with pancreatic head
  cancer after PPPD or Whipple procedure
   – Still very poor due to recurrent diseases
 Awaiting more research in this area of
 Role of extended lymphadenectomy ?

 Original hypothesis
   – radical pancreatic resection as a means of increasing
     resectability rates and improving the outcome for

 Rationale
   – Largely found on the patterns of failure after Whipple
     resection as a result of local intra-abdominal recurrence
   – removing all peri-pancreatic tissues and LNs that carry
     tumour cells should translate into better survival
Dr. Joseph Fortner’s 1973 illustration – total
pancreaticoduodenectomy and subtotal gastrectomy,
with arterial and venous resection (from Surgery
73:307-320; 1973)
Johns Hopkins prospective randomized
    study April 1996 to June 2001
 Johns Hopkins prospective randomized
     study April 1996 to June 2001

• 30-40% distal gastrectomy
      including LN stations 5 & 6 and some stations 3 & 4
      Including portions of greater omentum and lesser omentum along the
      course of right gastroepiploic artery and right gastric artery
• retroperitoneal dissection
      extended from right renal hilum to left lateral border of the aorta in the
      horizontal axis
      from portal vein to below the third part of duodenum in the vertical axis
      (LN station 16a2 and 16b1 + sampling of celiac LN [station 9]
Johns Hopkins prospective randomized
    study April 1996 to June 2001

  Actuarial survival curves for all patients (n=285; P=0.79)
      1-, 3-, and 5-year survival rates are 80%, 44% and 23% for
      the standard group
      1-, 3-, and 5-year survival rates are 77%, 44% and 29% for
      the radical group
 Role of extended lymphadenectomy ?

 2 randomized trials
   – European lymphadenectomy study group (40 vs. 41
   – Johns Hopkins group (146 vs. 148 patients)
   – No survival benefit with extended lymphadenectomy
   – No substantial evidence for routine use of extended
     lymphadenectomy for pancreatic cancer

                  Pedrazzoli P et al., Ann Surg 1998;228:508-517.
                  Yeo CJ et al., Ann Surg 2002; 236:355-368.
    Other interventions that may affect
            surgical outcomes
 Total parental nutrition
   – Does more harm than good
 Continuous or cyclic enteral nutrition
   – Cyclic enteral feeding ass. earlier oral feeding
 Somatostatin and its analogues
   – Highly controversial
   – Results were very different between European and US
   – Differences in study design may account for the
     different outcomes
           Adjuvant therapy after PD

 Non-randomized studies in the 80s suggested
  some improvement in survival with adjuvant
 European study group for pancreatic cancer
  (ESPAC-1 trial)
   – rather complicated 2 x 2 factorial randomization
   – Observation vs. chemoradiation alone vs.
     chemotherapy alone vs. combination of the two
   – Overwhelming advantage for chemotherapy (5-year
     actuarial survival of 29.0%)

              Neoptolemos JP et al., NEJM 2004; 350:1200-1210.
Kaplan-Meier estimates of survival according to whether
or not patients received systemic chemotherapy
Does volume counts?
    Volume does count for resection of pancreatic cancer
                                                                                 Mortality rate (%)
Reference (Year)            Region        Study        No. of        No. of       High-       Low-
                                          period    Institutions   resections    volume      volume
                                                                                hospitals   hospitals
Studies from the USA
Lieberman et al. (1995)   New York      1984-1991       184          1972          5.5      11.8-18.9
Janes et al. (1996)       USA           1983-1990       978          2263          4.2         7.7
Gordon et al. (1995)      Maryland      1988-1993       39            501          2.2       13-19.1
Glasgow et al. (1996)     California    1990-1994       298          1705          3.5      6.9-14.1
Gordon et al. (1998)      Maryland      1984-1995       43           1093          1.8        14.2
Begg et al. (1998)        USA           1984-1993                     742          5.8        12.9
Birkmeyer et al. (1999)   USA           1992-1995      1246          7229          4.1      12.7-16.1

Studies from Europe
Neoptolemos et al.        UK            1976-1996       21           1026          5.9      8.3-27.6
Gouma et al. (2000)       Netherlands   1994-1998                    1126          0.9        16.0
NYCRIS[9] (2000)          Yorkshire,    1986-1994       17            130          7.8        28.6
Nordback et al. (2002)    Finland       1990-1994       33            374          3.6        12.9
                                The 30-day mortality and hospital mortality rates


                       30                                                                            *

  Mortality rate (%)



                            A   B   C     D      E       F         G       H       I       J     K   L   M


                                           30-day mortality rate       Hospital mortality arte

Figure 15: The unadjusted 30-day mortality and hospital
mortality rates after PD among the 13 HA hospitals.

                                                             *Indicates statistical significance
Any novel therapy?
Molecular and genetic studies of
      pancreatic cancer
 The resection rate of pancreatic cancer is increasing with
  better imaging modalities
 Pre-operative drainage has been showed to have no added
  advantage but many are still practicing it due to logistic
 PPPD has the same oncological clearance as standard
 Improved techniques in fashioning of PJ have lead to a
  reduction in leakage and mortality
 The expected survival benefit of extended lymphadenectomy
  has not been demonstrated
 Systemic chemotherapy improves survival after PD
 Overall prognosis is still bad
 More basic researches may help to improve the treatment

Shared By: