Comparative study between amputation of the rectum in the classic Lloyd-Davies position and in ventral decubitus

1 Santa Casa de Misericórdia, Department of Surgery, Discipline of Coloproctology, São Paulo, SP, Brazil. 2 ABC Medical School, Medicine Undergraduate, Santo André, SP, Brazil. Novais Comparative study between amputation of the rectum in the classic Lloyd-Davies position and in ventral decubitus. 2 Rev Col Bras Cir 45(5):e1643 When in Lloyd-Davies's position, Miles's surgery is done in one time with two concomitant surgical teams: one in the abdomen and one in the perineum, so that the patient remains in the supine position with 90o flexion of the thighs over the abdomen and also 90o flexion of the legs on the thighs, abducted in leg supports. Such positioning promotes shorter surgical time, with a consequent faster postoperative recovery, and avoids the risks associated with the ventral position related to the reduction of cardiac output and compression of body structures8.9. Recent studies suggest that the ventral position for excision of the rectum presents advantages such as better rectal dissection, less bleeding and less occurrence of iatrogenic perforations. On the other hand, surgical time increases due to the need to change the position of the patient and the non-synchronization of abdominal time with excision of the rectum. Nonetheless, it has benefits in terms of relapse and better cancer outcomes10. In view of these facts, we intended to evaluate the benefits and disadvantages of the ventral decubitus position compared with that of Lloyd-Davies for the perineal time of APA in patients operated at the Central Hospital of the Santa Casa de Misericórdia in São Paulo.


INTRODUCTION
C olon and rectum cancer is the third most common type of cancer in Brazil in men (5%) and the second in women (6.4%) 1 .The most common histopathological type is adenocarcinoma, in 96% of the cases.Other histological types are rare malignant neoplasms, some requiring specific therapeutic conducts 2,3 .
The singular anatomy of the rectum, its retroperitoneal location and the proximity of urogenital organs, autonomic nerves and anal sphincters, makes surgical access relatively complex.
In addition, the dissection must be precise in the anatomic planes, since the medial dissection for the endopelvic fascia that lines the mesorectum can compromise the patient with local recurrence, whereas the lateral dissection to the avascular anatomical space presents with risk of mixed autonomic nerves lesion, with impotence in men and bladder dysfunction in both genders 4 .
Abdominoperineal amputation (APA) of the rectum is advocated for tumors of the lower third of the rectum, anal canal, tumors with pelvic invasion, metastatic tumors, and those associated with inflammatory bowel diseases 5 .Among the APA techniques, Miles surgery involves a definite colostomy, closure of the sigmoid above the tumor, ligation of the inferior mesenteric vessels, closure of the anus, and removal of the specimen through the perineum, encompassing the anus.It is a large surgery, with important intraoperative complications (bleeding and hypovolemic shock) and postoperative ones (perineal abscess, urinary tract infection, sexual impotence and neurogenic bladder).The surgical mortality of APA is between 0% and 6.5% [5][6][7] .The perineal time of the APA can be performed with the patient in the Lloyd-Davies position or in the ventral position.

Original Article
Comparative study between amputation of the rectum in the classic Lloyd-Davies position and in ventral decubitus.

Novais
Comparative study between amputation of the rectum in the classic Lloyd-Davies position and in ventral decubitus.Nonetheless, it has benefits in terms of relapse and better cancer outcomes 10 .
In view of these facts, we intended to evaluate the benefits and disadvantages of the ventral decubitus position compared with that of Lloyd-Davies for the perineal time of APA in patients operated at the Central Hospital of the Santa Casa de Misericórdia in São Paulo.We performed statistical analysis with the software Statistical Package for Social Sciences (SPSS), v.13.0.In the descriptive analysis, we calculated the absolute and relative frequencies (n, percentage) for the qualitative variables and the summary measures (mean, median, standard deviation, minimum and maximum) for the quantitative variables.We used the Chi-square or Fisher's exact test, and t-Student or Mann-Whitney test.We adopted a significance level of 5%.

This work was approved by the Institutional
Ethics Committee with the following reference number: 1,252,153.

RESULTS
The sample consisted of 56 patients, who underwent rectal APA in Lloyd-Davis or ventral positions, the latter being used only from November

Novais
Comparative study between amputation of the rectum in the classic Lloyd-Davies position and in ventral decubitus.
In the comparison between anesthesia times and surgical time, we observed a mean of 6.3 hours for the Lloyd-Davies group.For the ventral group, we found a mean anesthetic time of 7.7 hours, the comparison between the two groups being statistically significant (p<0.001).Regarding surgical time, the Lloyd-Davies group presented a mean of 5.28 hours, and the ventral group, of 6.4 hours, also with statistical significance (p=0.006).
The variation in the pre and postoperative hemoglobin levels of patients who did not receive blood transfusion in the Lloyd-Davies group showed an average of 2.24g/dl.In the ventral group, the mean variation was 2.62g/dl, without statistical significance between the two groups (p=0.442).The variation of the pre and postoperative hematocrit of patients who did not receive blood transfusion in the Lloyd-Davies group had an average of 6.81%.
A negative value of the hematocrit difference was found only in one patient, probably due to laboratory error.In the ventral group, the mean variation was 7.54%, not being statistically significant between the two groups (p=0.602).
As for intraoperative blood transfusion, 14 patients from the Lloyd-Davies group received an average of 1.8 red cell units, with a median of 2,0.In the ventral group, three patients received transfusion, with an average of 3.6 units.The difference between the two groups was not statistically significant (p=0.682).By analyzing the amount of intraoperative volume received (crystalloids and colloids), we observed that the Lloyd-Davies group received an average of 4,432ml.
The analysis was statistically significant (p=0.021).
We observed intraoperative complications in 13 (23.2%)patients, 11 (29.7%) of the Lloyd-Davies group and two (10.5%) of the ventral group, and the difference between them was not statistically significant (p=0.181,Fisher's test).Among these complications, we observed rectum perforation during dissection in one (2.7%)patient in the Lloyd-Davies position and in one (5.3%)patient in the ventral position, not statistically significant (p=1.0,Fisher's test).We observed a urethral lesion in only one (  As to relapse, six (10.7%) patients had local recurrence, and five (8.9%), systemic recurrence.Among the patients with local recurrence, five (13.5%) were of the Lloyd-Davies group, and one (5.3%), of the ventral group (p=0.65).We observed systemic recurrence in three (8.1%)patients of the Lloyd-Davies group and in two (10.5%) of the ventral group (p=1.0).For analysis of disease free time, we used a log-rank test, whose result was not statistically significant (p=0.193).
The overall survival time for the Lloyd-Davies group had a mean of 45.7 months, median of 46.1, ranging from one to 85.2 months, and a standard deviation of 22.9 months.For the ventral group, the mean was 15.5 months, median of 14.2, range of three to 33.4 months, and with standard deviation of 7.2.

DISCUSSION
The present study dealt with a sample with a predominance of rectal neoplasia cases in males, similar to the North American and global incidence, but unlike the Brazilian, which slightly favor women 1,[11][12][13][14] .The mean age at diagnosis of rectal neoplasia in our sample was compatible with those found in the literature, whose value is in the range of 63 years for men and 65 for women 14,15 .The predominant comorbidities of the sample were hypertension, diabetes mellitus and smoking.Several studies have linked diabetes with increased colorectal cancer, as well as smoking (especially for rectal cancer) [16][17][18][19][20][21][22] .The histological types observed are consistent with global statistics 2,3 .
Regarding the technical aspects of the comparison between the two surgical positions of interest for the abdominoperineal amputation of the rectum, we observed a statistically significant difference of the average surgical and anesthetic times, the Lloyd-Davies position having significantly faster execution in both features.Our data were contrary to other studies in the international literature, in which the ventral position was faster 23,24 .
As for intraoperative blood loss, analyzed by the difference in the hemoglobin and hematocrit levels in the pre and postoperative periods, we observed no statistically significant difference between the two positions.In the literature, the ventral position is associated with less blood loss than that of Lloyd-Davies 24 .Blood transfusions occurred in 37.8% of patients in the Lloyd-Davis group and in 15.8% of patients in the ventral group, although there was no statistically significant difference, such data being similar to international studies 24 .The difference in intraoperative intravenous volume received was statistically significant between the groups, the ventral group having received less.

Perioperative complications, early
complications and death did not show statistically significant differences between the groups, in contrast to the better performance of the ventral position in this aspect in the literature 24 .The analysis of late complications in general was statistically significant, with a greater predominance in the Lloyd-Davies group, in agreement with the researched literature 24 .
Regarding the duration of hospitalization, we found no statistical significance between the groups, but in the literature the ventral position displayed shorter hospital stay 24 .
The circumferential margins of the specimens, as well as the variation in the number of lymph nodes removed from the two groups, were also not statistically significant, as well as local and systemic recurrence.In some studies, local recurrence had a lower incidence in the ventral group, and in others, this difference was not identified 10,23 .The analysis of disease-free time showed no statistically significant difference between groups, a fact also observed in a Chinese study in 2015 24 .
We conclude that the analysis of the factors related to the different positions for the abdominoperineal amputation of the rectum revealed that the ventral position group had less need for intraoperative intravenous volume infusion and lower incidence of late postoperative complications, while the Lloyd-Davies position group obtained better surgical and anesthetic times.There were no statistically significant differences between the two groups in the bleeding and intraoperative blood transfusion rates, perioperative and early complications, deaths, duration of hospitalization, involvement of the circumferential margin and number of dissected lymph nodes.We shall perform the evaluation of relapse, disease-free time and overall survival after a longer follow-up.
Compared with the Lloyd-Davies position, the ventral position seems to be the best choice for patients undergoing abdominoperineal amputation of the rectum due to the lower rate of late complications.However, longer follow-up time is necessary to perform the choice of surgical position based on the oncologic outcome.

2
Rev Col Bras Cir 45(5):e1643When in Lloyd-Davies's position, Miles's surgery is done in one time with two concomitant surgical teams: one in the abdomen and one in the perineum, so that the patient remains in the supine position with 90º flexion of the thighs over the abdomen and also 90º flexion of the legs on the thighs, abducted in leg supports.Such positioning promotes shorter surgical time, with a consequent faster postoperative recovery, and avoids the risks associated with the ventral position related to the reduction of cardiac output and compression of body structures8.9.Recent studies suggest that the ventral position for excision of the rectum presents advantages such as better rectal dissection, less bleeding and less occurrence of iatrogenic perforations.On the other hand, surgical time increases due to the need to change the position of the patient and the non-synchronization of abdominal time with excision of the rectum.
We conducted a retrospective study of patients submitted to APA for distal rectal adenocarcinoma or squamous cell carcinoma of the anal canal, treated at the Central Hospital of Santa Casa de Misericórdia, São Paulo, between 2008 and 2018.This is a comparative analysis between the surgical tactics for the perineal time of rectal APA: Lloyd-Davies positioning versus ventral decubitus.Exclusion criteria were impossibility to obtain all the necessary data for analysis, surgery for amputation of rectum due to benign diseases, and patients who submitted only to a perineal time complementary to previous rectosigmoidectomy due to pelvic recurrence.We analyzed the variables age, gender, comorbidities, neoadjuvance, pathological staging, diagnosis, surgical time, need for crystalloids or transfusion of blood and intraoperative blood products, hospitalization time, circumferential and proximal anatomopathological margin, dissected lymph nodes, surgical staging, perioperative complications (shock, bleeding, injury of other organs, deep venous thrombosis, acute urinary retention), early postoperative (infection and surgical wound dehiscence, pelvic collection, intestinal subocclusion, colostomy ischemia) and late complications (paracolostomic hernia, urinary incontinence and sexual impotence), disease-free time, and overall survival.

( 2 .
7%) patient of the Lloyd-Davies group and in none of the ventral group (p=1.0).We observed paracolostomic hernia only in five (13.5%) patients of the Lloyd-Davies group (p=0.155,Fisher's test).We found subocclusion and acute obstructive abdomen in three patients, two (5.4%) of the Lloyd-Davies group and one (5.3%) of the ventral group (p=1.0).Regarding resection margins, the proximal was negative in all cases.The circumferential margin was positive in five cases, two (5.4%) of the Lloyd-Davies group and three (15.8%) of the ventral group, but not statistically significant (p=0.324,Fisher's test).We analyzed The number of resected lymph nodes with the Mann-Whitney test.The Lloyd-Davies group had a median of seven lymph nodes, ranging from zero to 25, and a mean of 8.51.In the ventral group, the median was 12 lymph nodes, with a range from zero to 23, and a mean of 11, without statistical significance (p=0.193).
2.7%) patient of the Lloyd-Davies group and in none of the other group.Intraoperative bleeding was reported for 11 patients, ten (27%) of the Lloyd-Davies group and one (5.3%) of the ventral group (p=0.77,Fisher's Test).