Page 4 - World Association of Laparoscopic Surgeons - Journal

RJ Orti-Rodríguez

outlines the search history and the total number of A weighted average (WA), utilized also by Gumbs et al
publications included in this review. in a recent review, is used to calculate a statistical weighted
The variables studied were as follows: Mean operating mean of all the differents means collected in the examined
room time, mean estimated blood loss, morbidity, bleeding, publications:
fistulas rates, mortality and conversion rates.
WA = (w x + w x +…+ w x )/(w + w +…+ w )
1
2
n
1 1
2 2
n n
RESULTS where w is the number of cases in a publication and x is
the mean of an specific variable.
The initial search identified 658 articles. After exclusion of
Tables 1 and 2 summarise the outcomes of total
repeated articles 228 and items not published in English or
laparoscopic and total robotic PD respectively.
Spanish languages or no humans studies, 44, 184 publications
were selected. All identified articles were examined, and PERIOPERATIVE FACTORS
manuscripts with one or more exclusion criteria were not
Mean operation time ranged from 287 to 628 minutes in
taken into account—review articles 68, single case reports
LG and from 312 to 718 minutes in RG, with a WA of
37, lack of relevance 62, data duplication 4.
389 (LG) and 394 minutes (RG). Mean estimated blood
During the process of data collection, we found an article
1
with a large series of 35 patients in which there were not loss ranged from 74 to 770 ml in LG, but this variable was
not reported in 10 cases (5%) which were not taken into
reported most of the variables examined, therefore, we
account in the analysis. Notice that the lowest estimated
finally excluded it in spite of it is mentioned and used in
blood loss is in the largest series as Palanivelu’s series and
2
other review articles. At the end, a total of 12 articles were
most of them are under 300 ml of blood loss. WA calculated
full examined.
was 178.7 ml. In the robotic arm WA was 319 ml with a
In the examined publications, there were no randomized
range from 153 to 389 ml. All of the groups reported
controlled trials and a total of 326 patients were included in
hospital stay. Average length of stay in LG ranged from
from all articles; 217 patients for the laparoscopic group
7 to 22.3 days with a WA of 10 days. The lowest limits of
(LG) and 109 for the robotic group (RG). In 25 patients
this range is from the group of Minnesota, the largest and
(LG), an additional assistance was necessary as a
one of the most recent series in this analysis. In RG, we
minilaparotomy or a hand-port to perform the whole
found a considerable increase in the length of stay, ranged
reconstruction, so these patients were excluded of the study.
from 9 to 22 days and with a WA of 15 and 31 days.
Of the remaining 301 patients the conversión rates were
29 (9.6%; LG = 16 (8.3%); RG = 13 (12%)) and were taken Morbidity
as technique fails.
Perioperative complications occurred in LG in 69 patients
(36%). We have stressed the importance of two specific
types of morbidity: Bleeding and fistulas and we have not
paid attention to gastric emptying because we are not going
to differentiate standard Whipple´s procedure with pylorus-
preserving PD. It is important to enhance that in RG total
morbidity is not representative because of the absence of
the largest series data (60 patients; 55%).
Bleeding
Bleeding is considerated as an unexpected blood loss during
the surgery or the whole postoperative hospital stay which
required any surgical or medical management and is not
directly derivated from another medical or nonmedical
intervention. Anemia as incidental finding was not
considerated as bleeding although blood transfusions were
required. Just one group did not report this data (42 patients;
26.4%) but we decided to count it as if they did not suffer
any bleeding because of the fact that the author reported a
detailled morbidity without the necessity of mention this
3
parameter. Bleeding was identified in eight (4%) of 192 of
Flow Chart 1: The search history and total number of
publications included the patients of the LG and in nine cases (9%) of 101 patients
50
JAYPEE