- Email: [email protected]
Experience with early postcesarean hospital dismissal
Experience with early postcesarean hospital dismissal Thomas H. Strong, Jr., MD: Willie L. Brown, Jr., MD,b Willie L. Brown, MD,b and Charles M. Curry, MDb Phoenix, Arizona, and Fresno, California OBJECTIVE: Our purpose was to assess early postcesarean hospital dismissal. STUDY DESIGN: A retrospective review was performed of all women receiving cesarean delivery over the most recent 6-month period in a busy private obstetrics practice that routinely dismisses its cesarean patients on postoperative day 2. Women who meet certain criteria (uncomplicated pregnancy, Pfannenstiel incision, uncomplicated surgery, no febrile morbidity, stable vital signs, ability to ambulate without aSSistance, ability to urinate without assistance, and auscultation of active bowel sounds) on postoperative day 2 are dismissed from the hospital. Outcomes were compared against women undergoing cesarean delivery during the 6 months immediately before the institution of the early dismissal program. RESULTS: Among 147 women undergoing cesarean deliveries, 117 (80%) met the criteria for early dismissal. When compared with controls (n = 93), there was no difference in outcomes. No one in the early dismissal group required readmission to the hospital. CONCLUSION: Among properly selected candidates, early postcesarean hospital admission is a reasonable option. (AM J OBSTET GVNECOL 1993;169:116-9.)
Key words: Early dismissal, cesarean delivery
Since the early 1940s, when women were hospitalized for 7 to 10 days after an uncomplicated vaginal delivery, the postpartum "lying-in" period has been significantly abbreviated. Today in Europe and the United States some obstetricians permit selected patients to go home within 12 hours of an uncomplicated vaginal delivery. In almost every instance where shorter in-hospital convalescence has been permitted, the impetus to change has arisen not out of medical enlightenment but as a result of the economic and logistic needs of the hospital. After World War II, for example, progressively shorter in-hospital puerperal stays were encouraged because of the overcrowding of maternity wards brought on by the "baby-boom." I With refinements in surgical technology and postoperative care the convalescent period after an uncomplicated cesarean delivery has also been shortened. Recently, for example, the average postcesarean stay at many maternity hospitals has fallen to 3 days.2 It has been our experience that the vast m
From the Phoenix Perinatal Associates, Division of Maternal-Fetal Medicine, Good Samaritan Regional Medical Center;' and the w.L. Brown Medical Corporation b . Received for publication March 26, 1992; revised December 1, 1992; accepted january 25,1993. Reprints not available. Copyright © 1993 by Mosby-Year Book, Inc.
0002-9378/93 $1.00 + .20 6/1/45895
116
patient requires the most intensive postoperative care. Since July 1989 we (W.L.B., W.L.B., Jr., C.M.C.) have followed a policy of routine, early postcesarean hospital admission. We present a report of our most recent expenence. Methods
The delivery and postpartum records of all women undergoing cesarean delivery between Jan. 1 and June 30, 1991, at a large, private obstetrics and gynecology practice in Fresno, California, were reviewed. The women served by this practice are of middle and lower socioeconomic status. All pregnant women entering the practice are informed that if they meet certain criteria they may be dismissed from the hospital on the second postoperative day (i.e., after 48 hours) should they undergo a cesarean delivery. During the third trimester the possibility of early postcesarean dismissal is again discussed with the patients during prenatal visits. The criteria for early hospital dismissal include: (1) an obstetrically and medically uncomplicated pregnancy, (2) a Pfannenstiel incision that is intact and free of inflammation or discharge on postoperative day 2. (At the discretion of the surgeon, skin staples may be left in place until postoperative day 3 or 4.), (3) no intraoperative complications (i.e., intraoperative blood loss > 1200 ml, bowel or urinary tract injury, etc.) or surgical drains, (4) absence of postpartum febrile morbidity (temperature;;:: 38.0° C), (5) stable vital signs and the ability to ambulate without assistance, (6) ausculta-
Volume 169, Number 1 Am J Obstet Gynecol
Strong et al.
117
Table I. Patient characteristics Early dismissal (n Age (yr) Primiparous Ruptured membranes Gestational age (wk)
=
117)
Control (n
24.7 ± 5.8 38 (32%) 51 (44%) 38.7 ± 1.7
=
Significance
93)
NS
23.8 ± 5.8 18 (19%) 43 (46%) 37.8 ± 5.8
P=
0.03*
NS NS
NS, Not significant. *x 2 analysis.
Table II. Characteristics of surgery Early dismissal (n
=
117)
Control (n
=
93)
p=
No. of prior cesarean sections
o 1
2 3 3t
Indication Failure to progress or cephalopelvic disproportion Repeat Distress Malpresentation Other Anesthesia General Regional Prophylactic antibiotics
Significance
61 (52%) 42 (36%) 9 (8%) 5 (4%)
32 (34%) 34 (37%) 17 (18%) 7 (8%) 3 (3%)
34 (30%)
18 (19%)
50 (43%) 6 (5%) 21 (18%) 5 (4%)
56 (60%) 7 (8%) 4 (4%) 8 (9%)
12 (10%) 105 (90%) 117 (100%)
15 (16%) 78 (84%) 93 (100%)
o
0.01*
NS
p=
0.01*
NS
P=
0.002t
NS NS NS NS
NS, Not significant. *X 2 analysis. tPearson's coefficient.
tion of normal bowel sounds by postoperative day 2, (7) the ability to urinate without assistance or catheterization, and (8) desire or willingness for early dismissal. In the event that any of the above criteria were not met, the patients were excluded from the study and remained hospitalized as dictated by their particular obstetric or medical condition. The control group consisted of all women undergoing cesarean delivery during the 6 months Oanuary 1 through June 30. 1989) immediately before the initiation of the early dismissal protocol (i.e., patients routinely dismissed on postoperative day 3 or later). Irrespective of the length of postpartum hospitalization, all patients were given explicit instructions regarding the signs and symptoms of the common postoperative complications.
Results A total of 147 women underwent cesarean delivery during the study period. All were delivered at the same hospital and all records were successfully retrieved for review. Within this group 117 (80%) met the inclusion criteria and agreed to be sent home on postoperative day 2. Ninety-three women underwent cesarean delivery in the control group. Table I lists the characteristics
of the subjects in the study and control groups. The indications for cesarean delivery are listed in Table II. The early dismissal group had significantly more primiparas (p = 0.03). There were significantly fewer women with prior cesarean deliveries (p = 0.01) and more with fetal malpresentations (p = 0.002) in the early dismissal group. The average duration of hospitalization in the early dismissal group was 44.3 ± 8.4 hours, whereas the mean duration of hospitalization for the control group was 70.8 ± 16.8 hours, a difference of 26.5 hours (p < 0.00005). The entire population had an overall postoperative complication rate of 6.2% (13 of 210). Table III compares the postoperative complications in the early dismissal and control groups. There was no significant difference in the incidence of postoperative complications between the two groups. Of the postoperative complications that arose in the early dismissal group, five (71 %) occurred 4 or more days after surgery. Similarly, 67% of the postoperative problems that developed in the control group occurred after postoperative day 4. None of the patients in the early dismissal group who experienced postoperative complications required readmission to the hospital. The early dis-
118
Strong et al.
July 1993 Am J Obstet Gynecol
Table III. Postoperative characteristics
Postoperative complications Type of complication Febrile morbidity Wound Other Postoperative day of complication 1
2 3 4
4+ Length of hospital stay (hr) No. of postoperative office visits 1
2 3 Hospital readmission
Early dismissal (n = 117)
Control (n = 93)
Significance NS
7 (6%)
6 (6.5%)
(n = 7)
(n = 6)
4 (57%) 3 (43%)
2 (33%) 4 (67%)
(n = 7)
(n = 6)
o o
1 (16.5%) 1 (16.5%)
o
2 (29%) 1 (14%) 4 (57%) 44.3 ± 8.4 (n = II7) 58 (49.6%) 56 (47.8%) 3 (2.6%)
o
NS NS NS NS
o
o
o
4 (67%) 70.8 ± 16.8 (n = 93) 86 (93%) 7 (7%)
p<
P<
0.00005* 0.00005t
o
1 (1%)
NS
NS, Not significant. *Student t test. t X2 analysis.
missal group required more postpartum office VISitS. However, this was largely for skin staple removal. Among those in the early dismissal group (n = 117) the positive predictive value (of meeting the early discharge criteria) for an uncomplicated postoperative course was 94%, whereas the sensitivity and specificity of the criteria were 92% and 75%, respectively. Comment
Common to most Western societies is the notion that puerperal care should be rendered in a hospital setting, especially if a cesarean delivery has occurred. When compared with the woman who is delivered vaginally, the cesarean patient has an increased risk for a number of complications, especially febrile morbidity. Nevertheless, the vast majority of women who undergo cesarean deliveries are healthy, well-nourished individuals who tolerate surgery well. In most cases the indications cited for cesarean delivery relate to an obstetric rather than a medical problem (i.e., protracted labor, malpresentation, etc.). Therefore postoperative recovery is usually quite rapid when compared with the ill or moribund patient. Among most women who undergo a cesarean delivery, hospital dismissal depends on the absence of febrile morbidity and the return of normal bowel function. In 90% of patients postoperative febrile morbidity is caused by atelectasis. After a cesarean delivery atelectasis rarely develops later than 48 hours. Epidural anesthesia and Pfannenstiel incisions, common among cesarean patients, may reduce postoperative impainnent of pulmonary function. Regarding postoperative return of bowel function, an array of factors may contribute to recovery among cesarean patients, including normal
serum electrolyte levels, a low rate of preoperative peritonitis, minimal bowel manipulation, and a relatively short duration of surgery. In short, the nature of the cesarean patient'S surgery may allow a more rapid recovery compared with other abdominal surgeries. It is noteworthy that among those in the early dismissal group who developed postoperative complications, most complications (71 %) occurred after the third postoperative day (i.e., the usual day of dismissal for many obstetric practices). Furthermore, 57% of the complications were noted at or beyond the fifth postoperative day (Table III). Had all 117 subjects from the study group remained hospitalized until the third postoperative day, only two additional patients (1.7%) with postoperative complications would have been identified. Interestingly, both patients weighed > 220 pounds and had body mass indices> 35. 3 The overall rate of postoperative complications (6.2%) compared favorably with previously published reports of postcesarean morbidity" However, the early dismissal group was at variance with the general obstetric population (and the control group) regarding the rate of fetal malpresentation. The reason for this is unclear. As a result of more fetal mal presentations and many women with prior cesarean deliveries (who incidentally requested repeat cesareans in spite of an active vaginal-birth-after-cesarean-delivery program), the early dismissal group contained many women who underwent cesarean deliveries after little or no labor. This may also have contributed to fewer postoperative complications. Given the relative infrequency of severe ileus, thromboembolic phenomena, and other uncommon postoperative problems, our study size was not large enough to assess their potential impact.
Strong et al.
Volume 169, Number 1 Am J Obstet Gynecol
We antiCipate that some may embrace this small, preliminary report to coerce patients and their obstetricians into a broad policy of early postcesarean dismissal. The authors recommend that early postcesarean hospital dismissal be considered a reasonable option and not a mandate at this time. On the basis of our data and experience, the following caveats are urged for the practitioner considering an early post-cesarean dismissal: (1) Strict adherence to the selection criteria should be observed. The morbidly obese patient should be considered a suboptimal candidate. (2) Ample antepartum preparation and psychoprophylaxis of the patient should occur; she should be a willing candidate. Until further studies are completed, it may be prudent to document that the patient is agreeable to early dismissal and that she is aware that in-hospital convalescence until postoperative day 3 is still an option. Not all women will choose early dismissal. (3) The patient'S home situation and emotional status should be considered. Although home health care was not included in this report, its use would likely be beneficial for the patient who is dismissed early. Even with home health visits, it is likely that less expense will be generated when compared with 3 or 4 days of hospitalization. We surveyed the postcesarean hospital charges at several institutions in the Fresno and Phoenix areas and found a range of $775 to $1170 per day. Assuming that 25% of the four million births that occur annually in the
119
United States are cesarean deliveries, one million women would be potential candidates for early dismissal. If 80% were to meet early dismissal criteria, as occurred in our group, a savings of $620 to $936 million would be realized annually. An advantage that the obstetrician has over many other surgical specialists is the relatively long patientdoctor relationship that precedes most surgical interventions. By using the antepartum period to familiarize the pregnant woman with the various obstetric contingencies, the obstetrician may have a favorable impact on the intrapartum and postpartum course, irrespective of delivery route. In our experience it appears that a patient's expectations playa significant role in determining when she goes home from the hospital. Therefore the importance of antepartum preparation of the early dismissal candidate cannot be overemphasized. REFERENCES 1. Schipani D. The new mom's hospital stay: is it over too soon? Child 1991;6:60-3. 2. Gillerman H, Beckham MH. The postpartum early discharge dilemma: an innovative solution. ] Perinat Neonat Nurs 1991;5:9-15. 3. Institute of Medicine (United States) Subcommittee on Nutritional Status and Weight Gain During Pregnancy. Nutrition during pregnancy. Washington: National Academy Press, 1990:433. 4. Nielsen TF, Hokegard KH. Postoperative cesarean section morbidity: a prospective study. AM] OBSTET GYNECOL 1983; 146:911-6.
Key words: Early dismissal, cesarean delivery
Since the early 1940s, when women were hospitalized for 7 to 10 days after an uncomplicated vaginal delivery, the postpartum "lying-in" period has been significantly abbreviated. Today in Europe and the United States some obstetricians permit selected patients to go home within 12 hours of an uncomplicated vaginal delivery. In almost every instance where shorter in-hospital convalescence has been permitted, the impetus to change has arisen not out of medical enlightenment but as a result of the economic and logistic needs of the hospital. After World War II, for example, progressively shorter in-hospital puerperal stays were encouraged because of the overcrowding of maternity wards brought on by the "baby-boom." I With refinements in surgical technology and postoperative care the convalescent period after an uncomplicated cesarean delivery has also been shortened. Recently, for example, the average postcesarean stay at many maternity hospitals has fallen to 3 days.2 It has been our experience that the vast m
From the Phoenix Perinatal Associates, Division of Maternal-Fetal Medicine, Good Samaritan Regional Medical Center;' and the w.L. Brown Medical Corporation b . Received for publication March 26, 1992; revised December 1, 1992; accepted january 25,1993. Reprints not available. Copyright © 1993 by Mosby-Year Book, Inc.
0002-9378/93 $1.00 + .20 6/1/45895
116
patient requires the most intensive postoperative care. Since July 1989 we (W.L.B., W.L.B., Jr., C.M.C.) have followed a policy of routine, early postcesarean hospital admission. We present a report of our most recent expenence. Methods
The delivery and postpartum records of all women undergoing cesarean delivery between Jan. 1 and June 30, 1991, at a large, private obstetrics and gynecology practice in Fresno, California, were reviewed. The women served by this practice are of middle and lower socioeconomic status. All pregnant women entering the practice are informed that if they meet certain criteria they may be dismissed from the hospital on the second postoperative day (i.e., after 48 hours) should they undergo a cesarean delivery. During the third trimester the possibility of early postcesarean dismissal is again discussed with the patients during prenatal visits. The criteria for early hospital dismissal include: (1) an obstetrically and medically uncomplicated pregnancy, (2) a Pfannenstiel incision that is intact and free of inflammation or discharge on postoperative day 2. (At the discretion of the surgeon, skin staples may be left in place until postoperative day 3 or 4.), (3) no intraoperative complications (i.e., intraoperative blood loss > 1200 ml, bowel or urinary tract injury, etc.) or surgical drains, (4) absence of postpartum febrile morbidity (temperature;;:: 38.0° C), (5) stable vital signs and the ability to ambulate without assistance, (6) ausculta-
Volume 169, Number 1 Am J Obstet Gynecol
Strong et al.
117
Table I. Patient characteristics Early dismissal (n Age (yr) Primiparous Ruptured membranes Gestational age (wk)
=
117)
Control (n
24.7 ± 5.8 38 (32%) 51 (44%) 38.7 ± 1.7
=
Significance
93)
NS
23.8 ± 5.8 18 (19%) 43 (46%) 37.8 ± 5.8
P=
0.03*
NS NS
NS, Not significant. *x 2 analysis.
Table II. Characteristics of surgery Early dismissal (n
=
117)
Control (n
=
93)
p=
No. of prior cesarean sections
o 1
2 3 3t
Indication Failure to progress or cephalopelvic disproportion Repeat Distress Malpresentation Other Anesthesia General Regional Prophylactic antibiotics
Significance
61 (52%) 42 (36%) 9 (8%) 5 (4%)
32 (34%) 34 (37%) 17 (18%) 7 (8%) 3 (3%)
34 (30%)
18 (19%)
50 (43%) 6 (5%) 21 (18%) 5 (4%)
56 (60%) 7 (8%) 4 (4%) 8 (9%)
12 (10%) 105 (90%) 117 (100%)
15 (16%) 78 (84%) 93 (100%)
o
0.01*
NS
p=
0.01*
NS
P=
0.002t
NS NS NS NS
NS, Not significant. *X 2 analysis. tPearson's coefficient.
tion of normal bowel sounds by postoperative day 2, (7) the ability to urinate without assistance or catheterization, and (8) desire or willingness for early dismissal. In the event that any of the above criteria were not met, the patients were excluded from the study and remained hospitalized as dictated by their particular obstetric or medical condition. The control group consisted of all women undergoing cesarean delivery during the 6 months Oanuary 1 through June 30. 1989) immediately before the initiation of the early dismissal protocol (i.e., patients routinely dismissed on postoperative day 3 or later). Irrespective of the length of postpartum hospitalization, all patients were given explicit instructions regarding the signs and symptoms of the common postoperative complications.
Results A total of 147 women underwent cesarean delivery during the study period. All were delivered at the same hospital and all records were successfully retrieved for review. Within this group 117 (80%) met the inclusion criteria and agreed to be sent home on postoperative day 2. Ninety-three women underwent cesarean delivery in the control group. Table I lists the characteristics
of the subjects in the study and control groups. The indications for cesarean delivery are listed in Table II. The early dismissal group had significantly more primiparas (p = 0.03). There were significantly fewer women with prior cesarean deliveries (p = 0.01) and more with fetal malpresentations (p = 0.002) in the early dismissal group. The average duration of hospitalization in the early dismissal group was 44.3 ± 8.4 hours, whereas the mean duration of hospitalization for the control group was 70.8 ± 16.8 hours, a difference of 26.5 hours (p < 0.00005). The entire population had an overall postoperative complication rate of 6.2% (13 of 210). Table III compares the postoperative complications in the early dismissal and control groups. There was no significant difference in the incidence of postoperative complications between the two groups. Of the postoperative complications that arose in the early dismissal group, five (71 %) occurred 4 or more days after surgery. Similarly, 67% of the postoperative problems that developed in the control group occurred after postoperative day 4. None of the patients in the early dismissal group who experienced postoperative complications required readmission to the hospital. The early dis-
118
Strong et al.
July 1993 Am J Obstet Gynecol
Table III. Postoperative characteristics
Postoperative complications Type of complication Febrile morbidity Wound Other Postoperative day of complication 1
2 3 4
4+ Length of hospital stay (hr) No. of postoperative office visits 1
2 3 Hospital readmission
Early dismissal (n = 117)
Control (n = 93)
Significance NS
7 (6%)
6 (6.5%)
(n = 7)
(n = 6)
4 (57%) 3 (43%)
2 (33%) 4 (67%)
(n = 7)
(n = 6)
o o
1 (16.5%) 1 (16.5%)
o
2 (29%) 1 (14%) 4 (57%) 44.3 ± 8.4 (n = II7) 58 (49.6%) 56 (47.8%) 3 (2.6%)
o
NS NS NS NS
o
o
o
4 (67%) 70.8 ± 16.8 (n = 93) 86 (93%) 7 (7%)
p<
P<
0.00005* 0.00005t
o
1 (1%)
NS
NS, Not significant. *Student t test. t X2 analysis.
missal group required more postpartum office VISitS. However, this was largely for skin staple removal. Among those in the early dismissal group (n = 117) the positive predictive value (of meeting the early discharge criteria) for an uncomplicated postoperative course was 94%, whereas the sensitivity and specificity of the criteria were 92% and 75%, respectively. Comment
Common to most Western societies is the notion that puerperal care should be rendered in a hospital setting, especially if a cesarean delivery has occurred. When compared with the woman who is delivered vaginally, the cesarean patient has an increased risk for a number of complications, especially febrile morbidity. Nevertheless, the vast majority of women who undergo cesarean deliveries are healthy, well-nourished individuals who tolerate surgery well. In most cases the indications cited for cesarean delivery relate to an obstetric rather than a medical problem (i.e., protracted labor, malpresentation, etc.). Therefore postoperative recovery is usually quite rapid when compared with the ill or moribund patient. Among most women who undergo a cesarean delivery, hospital dismissal depends on the absence of febrile morbidity and the return of normal bowel function. In 90% of patients postoperative febrile morbidity is caused by atelectasis. After a cesarean delivery atelectasis rarely develops later than 48 hours. Epidural anesthesia and Pfannenstiel incisions, common among cesarean patients, may reduce postoperative impainnent of pulmonary function. Regarding postoperative return of bowel function, an array of factors may contribute to recovery among cesarean patients, including normal
serum electrolyte levels, a low rate of preoperative peritonitis, minimal bowel manipulation, and a relatively short duration of surgery. In short, the nature of the cesarean patient'S surgery may allow a more rapid recovery compared with other abdominal surgeries. It is noteworthy that among those in the early dismissal group who developed postoperative complications, most complications (71 %) occurred after the third postoperative day (i.e., the usual day of dismissal for many obstetric practices). Furthermore, 57% of the complications were noted at or beyond the fifth postoperative day (Table III). Had all 117 subjects from the study group remained hospitalized until the third postoperative day, only two additional patients (1.7%) with postoperative complications would have been identified. Interestingly, both patients weighed > 220 pounds and had body mass indices> 35. 3 The overall rate of postoperative complications (6.2%) compared favorably with previously published reports of postcesarean morbidity" However, the early dismissal group was at variance with the general obstetric population (and the control group) regarding the rate of fetal malpresentation. The reason for this is unclear. As a result of more fetal mal presentations and many women with prior cesarean deliveries (who incidentally requested repeat cesareans in spite of an active vaginal-birth-after-cesarean-delivery program), the early dismissal group contained many women who underwent cesarean deliveries after little or no labor. This may also have contributed to fewer postoperative complications. Given the relative infrequency of severe ileus, thromboembolic phenomena, and other uncommon postoperative problems, our study size was not large enough to assess their potential impact.
Strong et al.
Volume 169, Number 1 Am J Obstet Gynecol
We antiCipate that some may embrace this small, preliminary report to coerce patients and their obstetricians into a broad policy of early postcesarean dismissal. The authors recommend that early postcesarean hospital dismissal be considered a reasonable option and not a mandate at this time. On the basis of our data and experience, the following caveats are urged for the practitioner considering an early post-cesarean dismissal: (1) Strict adherence to the selection criteria should be observed. The morbidly obese patient should be considered a suboptimal candidate. (2) Ample antepartum preparation and psychoprophylaxis of the patient should occur; she should be a willing candidate. Until further studies are completed, it may be prudent to document that the patient is agreeable to early dismissal and that she is aware that in-hospital convalescence until postoperative day 3 is still an option. Not all women will choose early dismissal. (3) The patient'S home situation and emotional status should be considered. Although home health care was not included in this report, its use would likely be beneficial for the patient who is dismissed early. Even with home health visits, it is likely that less expense will be generated when compared with 3 or 4 days of hospitalization. We surveyed the postcesarean hospital charges at several institutions in the Fresno and Phoenix areas and found a range of $775 to $1170 per day. Assuming that 25% of the four million births that occur annually in the
119
United States are cesarean deliveries, one million women would be potential candidates for early dismissal. If 80% were to meet early dismissal criteria, as occurred in our group, a savings of $620 to $936 million would be realized annually. An advantage that the obstetrician has over many other surgical specialists is the relatively long patientdoctor relationship that precedes most surgical interventions. By using the antepartum period to familiarize the pregnant woman with the various obstetric contingencies, the obstetrician may have a favorable impact on the intrapartum and postpartum course, irrespective of delivery route. In our experience it appears that a patient's expectations playa significant role in determining when she goes home from the hospital. Therefore the importance of antepartum preparation of the early dismissal candidate cannot be overemphasized. REFERENCES 1. Schipani D. The new mom's hospital stay: is it over too soon? Child 1991;6:60-3. 2. Gillerman H, Beckham MH. The postpartum early discharge dilemma: an innovative solution. ] Perinat Neonat Nurs 1991;5:9-15. 3. Institute of Medicine (United States) Subcommittee on Nutritional Status and Weight Gain During Pregnancy. Nutrition during pregnancy. Washington: National Academy Press, 1990:433. 4. Nielsen TF, Hokegard KH. Postoperative cesarean section morbidity: a prospective study. AM] OBSTET GYNECOL 1983; 146:911-6.