Prophylactic topical cefamandole in radical hysterectomy

133

Int. .1. C;ynuecol. Obsret., 1087. 2s: 133-138 International Federation of Gynaecology & Obstetrics

PROPHYLACTIC

K. MIYAZAWA,

(Received (Revision (Accepted

E. HERNANDEZ

of Obsterrics

Department

TOPICAL

CEFAMANDOLE

IN RADICAL

HYSTERECTOMY

and M.B. DILLON

and Gynecology,

Tripkr Army Medical

Center,

Honolulu,

HI 96859-5000

(USA)

May lSth, 1986) received August 8th, 1986) August 13th, 1986)

Abstract Miyazawa K, Hemandez E, Dillon MB (Department of Obstetrics and Gynecology, Tripler Army Medical Center, Honolulu, HI 96859, USA). Prophylactic topical cefamandole in radical hysterectomy. Int J Gynaecol Obstet 25: 133-138, 1987 From July 1, 1978 to June 30, 1984, 45 radical abdominal hysterectomies were performed by the authors at Tripler Army Medical Center. Management was uniform except for the use of prophylactic antibiotics. Three patterns of practice were identified: Group I, no antibiotics were used; Group II, intravenous (i.v.) antibiotics were given in the induction room and for less than 48 h postsurgery; Group III, prophylactic i.v. antibiotics were given and the surgical site was also im’gated with a cefamandole and saline solution. The three groups were found to be similar with regard to age, parity, weightheight index, pre- and postoperative hemutocrit, pre-operative white blood cell count, operative and anesthesia times, estimated blood loss, and amount of blood transfused. Groups I and II had a higher surgical site infection rate (87.5% and 63.6%, respectively) than Group III (3.8%). The mean 1O-day fever index in degree hours was 109 The opinions and assertions in this article are those of the authors and they are not to be interpreted as official views of the Department of the Army or the Department of Defense.

OO2f.h7292/87/$03.50 @ 1987 International Published and Printed

for Group I, 71 for Group II, and 30 for Group III (P < 0.001). Irrigation of the surgical site with a cefamandole and saline solution in addition to iv. an tibia tics decreases the infectious morbidity of radical hysterectomy.

Keywords: Hysterectomy; Infection; Antibiotics; Cephalosporins; Cancer; Surgery. Introduction Prophylactic antibiotics have been successfully employed in several obstetric and gynecologic procedures [ 11. Recently, prophylactic intravenous (i.v.) antibiotics have been shown to decrease febrile complications in patients undergoing radical abdominal hysterectomy [7,9]. In a prospective randomized blinded study, 34 patients undergoing radical abhysterectomy received either dominal 200 mg of doxycycline i.v. on call to the operating room, and 30 control patients did not [7]. The febrile morbidity was determined by the fever index in degree hours [5]. The doxycycline group had a statistically significant lower mean fever index. Vaginal cti and/or pelvic cellulitis developed in 26% of the control patients and in 12% of those receiving doxycycline. Sevin et al. [9] randomly assigned patients undergoing radical abdominal hysterectomy Int J Gynaecol

Federation in Ireland

of Gynaecology

& Obstetrics

Obster 25

I.34

Miyazawa

et al.

to receive either i.v. cefoxitin (2 g/dose) or placebo for a total of 12 doses. Each patient received three doses before surgery (every 6 h, starting 12 h before surgery), followed by nine doses after surgery (every 8 h after the last presurgical dose). Analysis of 53 patients completing the study revealed that 15% of the 26 patients receiving cefoxitin had surgical site-related infections compared with 52% of the 27 placebo patients (P = 0.005). In the present study, we evaluated the effectiveness of surgical site irrigation with a cefamandole and saline solution combined with i.v. cefamandole in preventing febrile morbidity from radical abdominal hysterectomy. Materials

and methods

From July 1, 1978 to June 30, 1984, 45 radical hysterectomies with pelvic lymphadenectomy were performed on the Gynecologic Oncology Service, Tripler Army Medical Center. Each of the operations was performed by one of the authors, all fully trained gynecologic oncologists, assisted by a senior resident or another Peri-operative management was author. uniform as it followed an established clinical protocol. Antibiotic prophylaxis was left to the discretion of the senior operating surgeon and this resulted in three groups of treatment patterns. In Group I, no antibiotics were used (8 patients); in Group II, i.v. antibiotics (cefamandole, 10 patients, and doxycycline, 1 patient) were given prophylactically immediately prior to and for less than 48 h after surgery; in Group III, prophylactic i.v. antibiotics (cefamandole, 25 patients, and doxycycline, 1 patient) were given together with cefamandole irrigation of the surgical site (Table I). All eight patients in group I and four patients in group II were operated by M.B.D.; six patients in group II and 18 in group III were operated by KM.; E.H. operated on one group II patient and eight group III patients. Int J Gynaecol

Obstet 25

Table 1. undergoing Group

Prophylactic antibiotics used radical abdominal hysterectomy.

I

Group

II

in

Group

45

patients

III

No antibiotics

IV antibiotics

IV antibiotics

N=8

Cefamandole N= 10 Doxycycline N=l

Cefamandole N=25 Doxycycline N=l

Saline irrigation

Saline irrigation

Cefamandole and saline irrigation

None of the 45 patients had an active infectious process prior to surgery. All patients underwent a 5-day mechanical and antibiotic bowel preparation, which included 8 g of neomycin given orally in divided doses the day before surgery and a l-l saline enema with 2 g of neomycin the night before surgery. They also received a vaginal douche with a povidone-iodine solution, and the abdomen was shaved. All patients received 5000 units of heparin subcutaneously 4 h before the operation and every 12 h thereafter until fully ambulatory. In the operating room, the perineum and vagina were prepared with a povidoneiodine solution. The abdomen was prepared with alcohol and painted with a povidoneiodine solution which was allowed to dry; then the abdomen was covered with sterile drapes. A transurethral catheter for gravity drainage was placed before the operation and removed on postoperative day 14. The surgical technique was the same in all cases. The abdomen was entered through a vertical midline incision from the symphisis to just above the umbilicus. After entering the peritoneal cavity, the abdomen was enlarged paraaortic or pelvic explored; nodes were removed and sent for frozen section. In the absence of enlarged nodes, or if the frozen section was negative, the pararectal and paravesical spaces were developed. The parametrium was palpated; in the absence of tumor, we proceeded with a radical hysterectomy. Briefly, the uterilie arteries were identified and ligated at their

Cefamondoie

origin. The ureters were mobilized laterally. The vesicovaginal and rectovaginal spaces were developed. All tissue medial to the origin of the uterine arteries was removed en bloc including a 3-S cm cuff of vagina. The vaginal cuff was closed in all cases. A pelvic lymphadenectomy was performed, starting above the bifurcation of the common iliacs and continuing caudal until reaching the inferior epigastric vessels. The lower common iliac, external iliac, and hypogastric nodes were removed, as were the nodes in the obturator fossae. Before closing the vaginal cuff, after completing the pelvic lymphadenectomy, and before closing the abdomen, the pelvis was irrigated with normal saline or a solution of 2g of cefamandole nafate in 1000 ml of normal saline. After the fascia was closed, the abdominal incision was also irrigated. The retroperitoneal spaces were drained with a closed suction drainage system. Drains were placed bilaterally in the retroperitoneum and exited through the skin of the lower abdominal quadrants. All patients had at least two sets of complete blood cell counts and serum electrolytes done postoperatively. Patients with a temperature between 38°C and 38.3”C had a complete physical examination with special attention to the lungs and abdomen, a complete blood cell count, urinalysis, and urine culture. If the temperature was above 38.3”C, the evaluation also included blood cultures, a pelvic examination with vaginal cuff cultures, and a chest radiograph when clinically indicated. Cultures of other possible sites of infection were taken when indicated. The febrile morbidity among the three patient groups was compared. Febrile morbidity was assessed by use of the fever index [S]. The fever index was expressed as degree hours and was calculated as the area, on the temperature graph, between a line extending horizontally at 37.2”C and a line joining adjacent temperature readings. Risk factors for febrile morbidity to in-

in radical hysterectomy

I35

clude age, parity, weight, height, preoperative hematocrit (Hct), postoperative lowest Hct, estimated blood loss, amount of blood transfused, operating time, anesthesia time, pre-operative white blood cell (WBC) counts and postoperative highest WBC counts were reviewed. For data analysis, the Student t-test or analysis of variance was used where appropriate. Results The mean age of the patients was 38.2 24-64). The racial comyears (range, position, was mixed and included Caucasians, Blacks, Polynesians, and Orientals. Cervical cancer was the indication for surgery in 43 patients. Thirty-three of these patients had squamous-cell carcinoma, nine had adenocarcinoma, and one had a carcinosarcoma. Thirty-five had stage 1 disease (30, IB; 5, IA) and eight had stage IIA. Stage II endometrial adenocarcinoma was the diagnosis in one case, and Stage I squamous-cell carcinoma of the vagina in another. There were no statistically significant differences in age, parity, weight-height index [3], blood loss, amount of blood transfused, pre-operative Hct, postoperative lowest Hct, operative time, anesthesia time and pre-operative WBC count among the three groups (Table II). The means of the lo-day fever index showed a statistically significant difference (P < 0.001). The mean lo-day fever index was lower (30 degree hours) for Group III than for Group II (71 degree hours), or Group I (109 degree hours). The means of the highest postoperative WBC counts were significantly different (P = 0.0004). Group I had the highest postoperative WBC count (16,062) compared to 10,763 for Group II and 11,888 for Group III. The Group II and Group III postoperative WBC counts were not statistically different. Infections were documented in eight patients in Group I, eight in Group II, and three in Group III (Table III). Surgical site hat J Gynarcol Obstet 25

136

Miyaznwa

et al.

Table 11. Characteristics

of the three study groups. Group N=8

I

Group II N= 11

Group N=26

III

Statistical analysis

Age (years)

Range Mean

26-62 42.5

27-66 41.5

24-64 38.1

NS

Parity

Range Mean

o-7 2.1

O-7 2.5

o-13 3.3

NS

Weight/height’

Range Mean

0.025-0.050 0.033

0.025-0.043 0.032

0.025-0.047 0.036

Range Mean

675-3400 2196

800-4000 1851

400-5500 1544

NS

Transfusion Amount (ml)

Range Mean

500-2500 1406

O-3000 1277

O-3500 1173

NS

Pre-operative Hematocrit (%)

Range Mean

32-42 37

2842 37

31-45 38

NS

Postoperative Hematocrit (%)

Range Mean

22-35 30

29-36 30

26-38 30

NS

Operative Time (min)

Range Mean

231-495 324

190-420 274

140-415 262

NS

Anesthesia Time (mitt)

Range Mean

274-508 362

227-461 323

166-440 300

NS

Pre-operative WBC/mm”

Range Mean

5700-10600 7837

5000-l 7154

4000-12300 7996

NS

Postoperative WBC/mm”

Range Mean

12000-22400 16062

7800-13500 10763

8300-17100 1188X

P = 0.0004

lo-day Index

Range Mean

64-213 109

26-l 71

2-80 30

Blood

loss (ml)

fever

NS = not significant

Table III. Type

14

Postoperative

infections

and Bacteriology Group N=8

cellulitis

results.

I

Group II N= I1

7 E. coli (5) Pseudomonas Enrerococcus

( 1)

5 E. coli (3) Pseudomonas

infection

1~’E. co/i (1)

2 Staphylococcus

Urinary

tract

1 E. coli (1)

1 Enrerococcus

Int J

infection vapinal

Gynaecol

cuti infection

Obstet 2.5

Group N=26

III

(2)

(1)

Wound

T‘oncurrcnt

P < 0.00 1

(P > 0.05).

of infection

Pelvic/cuff

1200

NS

(2)

(I )

2 E. cnli (2)

(tfamandole

infections occurred in 87.5% of Group I. 63.6%) of Group II. and 3.8% of Group III patients. Duration of hospitalization was not used as a measurement of morbidity because a high proportion of our patients were referred to us from other islands in the Pacific. In many instances, they stayed in the hospital past their recovery awaiting air transportation. Discussion

Radical abdominal hysterectomy is associated with a high degree of morbidity. In a recent report, 39 (74%) of 53 study patients dcvcloped postoperative morbidity [O]. Surgical site infections developed in 34% of these 53 patients. In our series, 15 (33%) of the 45 patients developed surgical site infections. A decrease in infectious morbidity from radical hysterectomy by using prophylactic i.v. antibiotics has been documented [7,9]. In one study a single pre-operative dose of doxycycline was used, while in the other study I2 peri-operative doses of cefoxitin were used. Creasman et al. [2] reported on the use of cefamandole in patients prophylactic undergoing extended pelvic surgery for adenocarcinoma of the endometrium. The patients received cefamandole, 2 g, or placebo i.v. on call to the operating room and every 6 h thereafter for a total of five doses. Febrile morbidity was more frequent in the placebo group. Irrigation of the surgical site with cefamandole was successfully used in our department in preventing postpartum endometritis in patients undergoing cessection investigators arean [S]. The theorized that delivering the antibiotic directly to the surgical site was important, as blood flow to the area may be restricted by hemostatic suturing and systemically administered

antibiotics

‘l‘opical prophylactic

may not be effective.

antibiotics have also

in rudiculhysterectomy

137

been successfully employed in non-gynecologic operations [4]. In a prospective, randomized, blinded study, 2 I7 patients undergoing a vascular procedure with a groin incision were allocated to one of four groups with respect to prophylactic antibiotics [6]. Group I received no antibiotics. Group II had topical cephradine instilled in their incisions before closure. Group III received a 24-h peri-operative course of i.v. cephradine, and Group IV received both topical and i.v. cephradine. Surgical site infections were significantly reduced (P -: 0.01) among patients in Groups II-IV. No significant differences were noted among the three antibiotic groups. In our study febrile morbidity, measured by the fever index, was significantly less in patients who received topical and i.v. antibiotics than in those receiving iv. antibiotics only (P = 0.0004). Although a decrease in infections and febrile morbidity was seen when i.v. antibiotics were used, a more drastic decrease was observed in patients receiving topical cefamandole. The data obtained from this non-randomized study presents some difficulties. There was not an equal number of patients in each of the study groups. Nevertheless, there were no significant differences among the groups with regards to risk factors. The differences in infection rate could be attributed to the different surgeons performing the procedures, even though the surgical technique, pre- and postoperative management were similar. Even after considering these negative points, the differences in infectious morbidity were so significant that use of topical cefamandole in radical abdominal hysterectomies seems justified. Further studies of other topical and parenteral prophylactic antibiotics should be considered. Conclusions

Infectious morbidity varied among 45 patients undergoing radical abdominal hys-

138

Miyazawn

et al.

terectomy with bilateral pelvic lymphadenectomy and was determined by the use and type of prophylactic antibiotics. Eight patients, Group I, received no antibiotics; Group II received i.v. cefamandole (10 patients), and doxycycline (1 patient) in the induction room and for less than 48-h postsurgery; Group III was given prophylactic i.v. antibiotics as above (cefamandole, 25 patients; doxycycline, 1 patient), and topical cefamandole was used. Febrile morbidity was measured by the fever index in degree hours [5]. The means of the lo-day fever index showed statistically significant differences (P < 0.001). The mean lo-day fever index was lower (30 degree hours) for Group III than for Group II (71 degree hours) and Group I (109 degree hours). Surgical site infections occurred in 87.5% of Group I, 63.6% of Group II, and 3.8% of Group III patients. These data demonstrate a role for topical cefamandole in reducing radical abdominal hysterectomy associated infections. References 1 Cartwright PS, Pittaway DE, Jones HW III, Entman SS: The use of prophylactic antibiotics in obstetrics and gynecology. A review. Obstet Gynecol Surv 39: 537, 1984.

Int .I Gynaecol

Obstet 25

2 Creasman WT. Hill GB, Weed JC Jr., et al.: A trial of prophylactic cefamandole in extended gynecologic surgery. Obstet Gynecol 59: 309, 1982. 3 Edwards KD, Whyte HM: The simple measurement of obesity. Clin Sci 22: 347. 1962. 4 Halasz NA: Wound infections and topical antibiotics. Arch Surg 112: 1240, 1977. 5 Ledger WJ, Kriewall TJ: The fever index: a quantitative indirect measure of hospital acquired infections in obstetrics and gynecology. Am J Obstet Gynecol 115: 5 14, 1973. 6 Pitt HA, Postier RG, MacGowan WAL et al.: Prophvlattic antibiotics in vascular surgery. Ann Surg ZY2:‘356. 1980. Rosenshein NB, Ruth JC, Villar J et al.: A prospective randomized study of doxycycline as a prophylactic antibiotic in patients undergoing radical hysterectomy. Gynecol Oncol 15: 201, 1983. Rudd EG, Cobey EA, Long WH et al.: Prevention of endoymyometritis using antibiotic irrigation during caesarean section. Obstet Gynecol 60: 413, 1982. Sevin, B-U, Ramos R, Lichtinger M et al.: Antibiotic prevention of infections complicating radical abdominal hysterectomy. Obstet Gynecol 64: 539, 1984.

Address for reprints: Colonel Kunfo Mfyazawa, M.D. P.O. Box 194 Trfpfer Army Medfcal Center Honofwfo, HI 968596000 U.S.A.