Objective
Elective cesarean section (CS) is a proven method to prevent mother-to-child transmission (MTCT), but is no longer recommended for women with antiretroviral therapy resulting in a low viral load (VL): <400 copies/mL in French and <1000 copies/mL in US guidelines. We sought to describe mode of delivery practices in human immunodeficiency virus (HIV)-infected women and their association with MTCT and postpartum complications.
Study Design
All deliveries from HIV-1-infected women in the French Perinatal Cohort (Agence Nationale de Recherches sur le Sida/Enquête Périnatale Française) 2000 through 2010 (N = 8977) were analyzed, with additional details for 2005 through 2010 (n = 4717).
Results
Vaginal deliveries increased from 25% in 2000 to 53% in 2010. Over 2005 through 2010, 4300 women had VL before delivery <400 copies/mL; among them only 49.3% delivered vaginally, 22.0% had nonelective CS, and 28.7% had elective CS. Elective CS were performed for scarred uterus in 45.4%, other obstetrical indications in 37.1%, and solely because of HIV in 15.7%. Of the 417 women with VL ≥400 copies/mL, 48.9% had elective CS as recommended, 25.9% had nonelective CS, and 25.2% had vaginal delivery. The MTCT rate did not differ according to the mode of delivery in term deliveries (≥37 gestational weeks) in 2000 through 2010: 0.3% after both vaginal delivery and elective CS with VL <50 copies/mL, 4.0% vs 5.3%, respectively, with VL ≥10,000 copies/mL. In case of preterm delivery, MTCT rates tended to be higher with vaginal delivery. Postpartum complications were more frequent following CS than vaginal deliveries (6.5% vs 2.9, P < .01).
Conclusion
Our findings suggest that HIV-infected women on antiretroviral therapy with low VL can safely opt for vaginal delivery in the absence of obstetrical risk factors.
A protective effect of elective cesarean section (CS) was proved in the absence of antiretroviral (ARV) therapy or with zidovudine monotherapy, leading to 2- to 5-fold reductions in mother-to-child transmission (MTCT) of human immunodeficiency virus (HIV) compared to vaginal delivery, both in large observational studies and in a randomized clinical trial. Elective CS at 38 gestational weeks was thus recommended for all HIV-infected women in all industrialized countries from 1997 through 1998. Subsequently, the use of combination ARV therapy (cART) was found to be the most effective means of preventing MTCT (PMTCT). Several observational studies suggested that there was no longer any additional benefit from elective CS beyond that conferred by cART and viral suppression, although some continued to show a protective effect.
Since 2002, French national guidelines recommend vaginal delivery for women with low viral loads (VLs) <400 copies/mL. However, no study was able to define a threshold below which elective CS has no benefit for MTCT. This has led to differences in clinical practice and national guidelines, where the VL threshold above which CS is recommended varies between countries: 400 copies/mL in France ; 1000 copies/mL in the United States, Canada, and Spain ; and 50 copies/mL in the United Kingdom as well as the European AIDS Clinical Society. Differences in mode of delivery practices in HIV-infected women between Europe and the United States were reported several years ago.
In France, although most HIV-infected women now achieve a VL <400 copies/mL before delivery, their rate of cesarean (elective and nonelective) remains well above that of the general population. This is a cause for concern because of the risks of CS, including short-term maternal and neonatal complications (and risks for future pregnancies including uterine rupture, placenta accreta, and repeat cesarean). The persistently high cesarean rate may be due to obstetricians and/or patients feeling reassured by avoiding vaginal delivery to avoid intrapartum transmission, even with a low VL.
Our objectives were to evaluate mode of delivery practices in France in the cART era with regard to: (1) the rate of MTCT of HIV, (2) how current guidelines were actually applied, and (3) the frequency of postpartum complications.
Materials and Methods
The ANRS French Perinatal HIV Cohort (EPF)
The nationwide Enquête Périnatale Française (EPF) has prospectively collected data on HIV-infected pregnant women and their children in centers throughout France since 1984, as detailed elsewhere. No specific recommendations for obstetric and HIV care were made, but investigators were encouraged to follow current French guidelines. Informed consent was obtained from the mothers, with an overall participation rate of 95%. This cohort was approved by the Hôpital Cochin Institutional Review Board and the French data protection agency ( Commission Nationale Informatique et Libertés ). Since 2005, EPF has been divided into a detailed cohort (EPF-CO1) in larger maternity units accounting for two-thirds of enrollment, and a less detailed cohort (EPF-CO11) in generally smaller maternity units.
Study population
The overall analysis was conducted among the 8977 pregnancies from women infected with HIV-1 delivering in participating hospitals from Jan. 1, 2000, through Dec. 31, 2010, among them 4745 in 2005 through 2010, after excluding 1276 women with missing data on mode of delivery or VL.
Variables studied
Main outcome was the mode of delivery, with 3 categories: vaginal delivery; elective CS, defined as before the onset of labor and intact membranes; and nonelective CS. Five types of indication for CS were recorded in the detailed EPF-CO1 since 2005: repeat cesarean (scarred uterus), premature rupture of membranes, other maternal or obstetrical indications, fetal indications, and PMTCT of HIV. Postpartum complications and hospitalization >7 days were also collected in the EPF-CO1.
Sociodemographic, obstetrical, and HIV treatment data included age, geographic origin (categorized as sub-Saharan Africa, metropolitan France, or other origins), gestational age (GA) at booking in the maternity, type of ARV therapy and time of treatment initiation and cessation, and GA at delivery, preterm birth being defined by GA <37 weeks. CD4 cell counts and plasma HIV-1 RNA VLs obtained closest to the time of delivery, not >7 days after delivery, were collected for all women. For some analyses, we considered only the last VL before delivery. Maternities were classified according to their perinatal care level: maternity only (level 1), onsite neonatology (level 2), and neonatal intensive care (level 3).
A child was considered infected if HIV-1 DNA or RNA polymerase chain reaction was positive on 2 samples or HIV-1 antibodies were detected at 18 months of age, or uninfected in other cases.
Statistical analysis
We first described modes of delivery from 2000 through 2010, according to whether the VL nearest to delivery was > or <400 copies/mL, the threshold to recommend elective CS in French guidelines. We then compared the MTCT rate according to mode of delivery, VL, and preterm birth in the subgroup of women treated with cART.
A detailed analysis of the mode of delivery was performed for the recent period 2005 through 2010, in which systematic cART was recommended (N = 4717). We identified factors associated with vaginal delivery separately according to VL categories (> or <400 copies/mL). We performed multivariate logistic regressions with mode of delivery (vaginal vs cesarean) as the dependent variable, adjusting for all noncollinear variables found to be associated with P < .20 in univariate analysis, using χ 2 or Fisher exact test for categorical variables, Student t or Wilcoxon test for continuous variables as appropriate. We then described the indications for CS in women in the EPF-CO1 with available data (N = 4001).
We also compared the frequency of postpartum complications and prolonged postpartum hospitalization (>7 days) according to mode of delivery and CD4 lymphocyte counts.
Results
Mode of delivery and transmission rates over the period 2000 through 2010
The rate of vaginal delivery increased from 25% in 2000 to 53% in 2010, with a parallel decrease in elective CS (57-26%), and a stable proportion of nonelective cesareans ( P for tendency < .0001) ( Figure ). This pattern mainly concerned the 7276 women with viral loads <400 copies/mL. For the 1701 women whose VL was ≥400 copies/mL, the rate of vaginal delivery increased slightly from 21% in 2000 to 29% in 2005 and remained quite stable thereafter.
In line with changes in French guidelines, the proportion of women who received cART during pregnancy increased from 41% in 2000, to 92% in 2005 and 98% in 2010. The proportion who received no ARV was 2% through the entire period. When considering term deliveries ≥37 weeks, rates of MTCT of HIV were similar with vaginal delivery and elective CS, whatever the VL. Results are presented by strata of VL ( Table 1 ). When the results were examined by setting thresholds at 50, 400, and 1000 copies/mL, there was no difference in MTCT according to mode of delivery between the different cutoffs. In preterm deliveries, the rates tended to be higher with vaginal delivery, but not significantly: 14.3% vs 6.3% for VL ≥10,000 copies/mL, P = .53 and 0.5% vs 0.0% with VL <50 copies/mL, P = 1.0.
| Variable | (1) Vaginal delivery | (2) Elective cesarean | P value (1) vs (2) | (3) Nonelective cesarean | P value (1) vs (3) | |||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| N | % | n | N | % | n | N | % | n | ||||
| Term deliveries (≥37 wk) | According to VL categories, copies/mL a | |||||||||||
| ≥10,000 | 25 | 4.0 | (1) | 113 | 5.3 | (6) | 1.00 | 31 | 9.7 | (3) | .41 | |
| 1000-10,000 | 52 | 1.9 | (1) | 160 | 0.6 | (1) | .43 | 49 | 0.0 | (0) | 1.00 | |
| 400-1000 | 37 | 0.0 | (0) | 100 | 0.0 | (0) | NA | 31 | 0.0 | (0) | NA | |
| 50-400 | 307 | 1.0 | (3) | 400 | 1.0 | (4) | 1.00 | 159 | 2.5 | (4) | .24 | |
| <50 | 1880 | 0.3 | (6) | 1195 | 0.3 | (3) | 1.00 | 703 | 0.3 | (2) | 1.00 | |
| Preterm deliveries (<37 wk) | According to VL categories, copies/mL a | |||||||||||
| ≥10,000 | 7 | 14.3 | (1) | 16 | 6.3 | (1) | .53 | 26 | 0.0 | (0) | .21 | |
| 1000-10,000 | 11 | 0.0 | (0) | 24 | 0.0 | (0) | NA | 25 | 8.0 | (2) | 1.00 | |
| 400-1000 | 9 | 11.1 | (1) | 17 | 5.9 | (1) | 1.00 | 14 | 0.0 | (0) | .39 | |
| 50-400 | 39 | 2.6 | (1) | 56 | 1.8 | (1) | 1.00 | 82 | 3.7 | (3) | 1.00 | |
| <50 | 189 | 0.5 | (1) | 143 | 0.0 | (0) | 1.00 | 201 | 0.5 | (1) | 1.00 | |
a Cases with undetectable VL were excluded when detection limit of human immunodeficiency virus-1 RNA assay used was >50 copies/mL (n = 440).
Factors associated with mode of delivery over the period 2005 through 2010
Among the 417 women whose last VL before delivery was ≥400 copies/mL, 25.2% delivered vaginally, compared with 49.3% of the 4300 women having a VL <400 copies/mL ( P < .01).
In the group with VL ≥400 copies/mL, vaginal delivery increased with timely management (booking maternity, HIV diagnosis, and ARV initiation before the third trimester), no severe immune deficiency (CD4 ≥200 cell/μL), as well as in very preterm compared with term delivery, and level 1 vs level 2 and 3 maternities ( Table 2 ). Early booking, CD4 cell count ≥200 cell/μL, delivery <32 gestational weeks, and type of maternity remained significantly associated with vaginal delivery in multivariate analysis. PMTCT was the most common indication for elective CS (51.7% exclusively and 24.5% associated with other indication) ( Table 3 ). The other indications (nonexclusive) were repeat cesarean (27.8%), fetal indications (17.2%), and obstetrical and/or maternal reasons (14.6%). The main indications for nonelective cesareans, alone or combined, were PMTCT (65.7%) and premature rupture of membranes (38.6%).
| Variable | VL <400 copies/mL before delivery (n = 4300 d ) | VL ≥400 copies/mL before delivery (n = 417 d ) | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Univariate analysis | Multivariate analysis (n = 3595) | Univariate analysis | Multivariate analysis (n = 303) | |||||||||||||||
| N | % | (n) | OR c | (95% CI) | P value | OR a | (95% CI) | P value | N | % | (n) | OR c | (95% CI) | P value | OR a | (95% CI) | P value | |
| Period, y | ||||||||||||||||||
| 2005–2006 | 1294 | 46.8 | (606) | 1 | < .01 | 1 | .07 | 157 | 23.6 | (37) | 1 | .77 | ||||||
| 2007–2008 | 1557 | 48.2 | (750) | 1.06 | (0.91–1.22) | 1.04 | (0.88–1.23) | 159 | 27.0 | (43) | 1.20 | (0.72–2.00) | ||||||
| 2009–2010 | 1449 | 52.9 | (766) | 1.27 | (1.10–1.48) | 1.20 | (1.02–1.42) | 101 | 24.8 | (25) | 1.07 | (0.60–1.91) | ||||||
| Level of perinatal center a | ||||||||||||||||||
| Level 1 | 220 | 56.4 | (124) | 1 | .03 | 1 | .08 | 34 | 41.2 | (14) | 1 | .03 | 1 | < .01 | ||||
| Levels 2-3 | 4037 | 49.0 | (1978) | 0.74 | (0.57–0.98) | 0.77 | (0.58–1.03) | 376 | 24.2 | (91) | 0.46 | (0.22–0.94) | 0.32 | (0.13–0.74) | ||||
| Location of perinatal center | ||||||||||||||||||
| Paris region | 3277 | 50.9 | (1669) | 1.31 | (1.14–1.51) | < .01 | 1.52 | (1.28–1.81) | < .01 | 290 | 27.6 | (80) | 1.45 | (0.87–2.41) | .16 | 1.27 | (0.56–2.91) | .57 |
| Other regions | 980 | 44.2 | (433) | 1 | 1 | 120 | 20.8 | (25) | 1 | 1 | ||||||||
| Geographic origin of mother | ||||||||||||||||||
| Metropolitan France | 637 | 46.9 | (299) | 1 | .36 | 60 | 18.3 | (11) | 1 | .35 | ||||||||
| Sub-Saharan Africa | 3166 | 49.8 | (1577) | 1.12 | (0.95–1.33) | 284 | 26.1 | (74) | 1.57 | (0.78–3.18) | ||||||||
| Other | 434 | 50.7 | (220) | 1.16 | (0.91–1.48) | 58 | 29.3 | (17) | 1.85 | (0.78–4.38) | ||||||||
| Marital status b | ||||||||||||||||||
| Single | 1342 | 49.9 | (669) | 1 | .12 | 140 | 31.4 | (44) | 1 | .07 | ||||||||
| Couple | 2209 | 49.6 | (1095) | 0.92 | (0.83–1.02) | 161 | 21.1 | (34) | 0.66 | (0.43–1.03) | ||||||||
| Booking visit in maternity | ||||||||||||||||||
| 1st-2nd trimester | 3730 | 49.6 | (1849) | 1.34 | (1.02–1.76) | .04 | 1.06 | (0.76–1.47) | .73 | 296 | 28.7 | (85) | 2.42 | (1.14–5.11) | .02 | 3.36 | (0.85–13.23) | .08 |
| 3rd trimester | 222 | 42.3 | (94) | 1 | 1 | 63 | 14.3 | (9) | 1 | 1 | ||||||||
| HIV diagnosis c | ||||||||||||||||||
| Before pregnancy | 3570 | 48.6 | (1736) | 1.66 | (0.81–3.38) | .07 | 1.30 | (0.61–2.78) | .05 | 298 | 23.5 | (70) | 2.66 | (0.78–9.05) | .04 | 0.92 | (0.21–4.06) | .72 |
| 1st-2nd trimester | 562 | 52.7 | (296) | 1.95 | (0.94–4.03) | 1.62 | (0.75–3.52) | 57 | 35.1 | (20) | 4.68 | (1.26–17.41) | 1.26 | (0.26–6.10) | ||||
| 3rd trimester | 33 | 36.4 | (12) | 1 | 1 | 29 | 10.3 | (3) | 1 | 1 | ||||||||
| HIV-1 RNA concentration before delivery, copies/mL | ||||||||||||||||||
| <50 | 3005 | 51.5 | (1547) | 1.65 | (1.40–1.96) | < .01 | 1.69 | (1.41–2.02) | < .01 | |||||||||
| ≥50 | 678 | 39.1 | (265) | 1 | 1 | |||||||||||||
| CD4 cell count before delivery, cells/μL | ||||||||||||||||||
| <200 | 197 | 36.6 | (72) | 1 | < .01 | 1 | < .01 | 59 | 13.6 | (8) | 1 | .02 | 1 | .08 | ||||
| ≥200 | 3467 | 50.0 | (1732) | 1.73 | (1.29–2.33) | 1.70 | (1.25–2.32) | 253 | 28.9 | (73) | 2.59 | (1.17–5.72) | 2.31 | (0.92–5.80) | ||||
| Parity | ||||||||||||||||||
| 0 | 1387 | 47.4 | (657) | 1 | .18 | 146 | 24.7 | (36) | 1 | .98 | ||||||||
| 1-2 | 2108 | 49.8 | (1050) | 1.10 | (0.96–1.26) | 194 | 25.3 | (49) | 1.03 | (0.63–1.70) | ||||||||
| ≥3 | 775 | 51.2 | (397) | 1.17 | (0.98–1.39) | 74 | 25.7 | (19) | 1.06 | (0.56–2.01) | ||||||||
| Twins | ||||||||||||||||||
| No | 4221 | 50.1 | (2106) | 5.21 | (3.04–8.93) | < .01 | 5.69 | (2.88–11.26) | < .01 | 403 | 25.3 | (102) | 1.24 | (0.34–4.54) | .74 | |||
| Yes | 99 | 16.2 | (16) | 1 | 1 | 14 | 21.4 | (3) | 1 | |||||||||
| Gestational age at delivery, wk | ||||||||||||||||||
| <32 | 144 | 46.5 | (67) | 0.84 | (0.60–1.17) | < .01 | 1.08 | (0.74–1.586) | < .01 | 45 | 73.3 | (33) | 10.82 | (5.28–22.17) | < .01 | 22.59 | (8.46–60.35) | < .01 |
| 32-36 | 475 | 38.1 | (181) | 0.59 | (0.49–0.72) | 0.68 | (0.54–0.84) | 64 | 14.1 | (9) | 0.64 | (0.30–1.37) | 0.99 | (0.42–2.35) | ||||
| 37-41 | 3671 | 50.9 | (1868) | 1 | 1 | 306 | 20.3 | (62) | 1 | 1 | ||||||||
a Level 1 have obstetrical unit only, level 2 have on-site neonatology, and level 3 have neonatal intensive care
b Data collected in Enquête Périnatale Française Cohort 1
d All study population; for women with undetectable VL, number of copies was equal to half of threshold.
| Variable | CS elective | CS nonelective | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| VL <400 copies/mL N = 1078 | VL ≥400 copies/m N = 161 | P value | VL <400 copies/mL N = 793 | VL ≥400 copies/mL N = 75 | P value | |||||
| % | (n) | % | (n) | % | (n) | % | (n) | |||
| History of CS | ||||||||||
| All | 45.4 | (472) | 27.8 | (42) | < .01 | 20.4 | (154) | 17.1 | (12) | .52 |
| Associated with PMTCT | 8.7 | (90) | 14.6 | (22) | 1.9 | (14) | 7.1 | (5) | ||
| Associated with other indications | 16.0 | (166) | 6.0 | (9) | 14.0 | (106) | 8.6 | (6) | ||
| Exclusively | 20.8 | (216) | 7.3 | (11) | 4.5 | (34) | 1.4 | (1) | ||
| Premature rupture of membranes | ||||||||||
| All | 0.6 | (6) | 1.3 | (2) | .29 | 38.6 | (292) | 38.6 | (27) | .99 |
| Associated with PMTCT | 0.0 | (0) | 0.0 | (0) | 4.9 | (37) | 15.7 | (11) | ||
| Associated with other indications | 0.50 | (5) | 0.7 | (1) | 27.5 | (208) | 20.0 | (14) | ||
| Exclusively | 0.1 | (1) | 0.7 | (1) | 6.2 | (47) | 2.9 | (2) | ||
| Other obstetrical or maternal indications a | ||||||||||
| All | 37.1 | (385) | 14.6 | (22) | < .01 | 47.0 | (355) | 24.3 | (17) | < .01 |
| Associated with PMTCT | 4.4 | (46) | 3.3 | (5) | 0.9 | (7) | 4.2 | (3) | ||
| Associated with other indications | 18.0 | (187) | 9.3 | (14) | 28.6 | (216) | 14.3 | (10) | ||
| Exclusively | 14.6 | (152) | 2.0 | (3) | 17.5 | (132) | 5.7 | (4) | ||
| Fetal indications b | ||||||||||
| All | 21.9 | (227) | 17.2 | (26) | .19 | 35.6 | (269) | 17.1 | (12) | < .01 |
| Associated with PMTCT | 3.1 | (32) | 3.3 | (5) | 0.3 | (2) | 1.4 | (1) | ||
| Associated with other indications | 9.8 | (102) | 8.6 | (13) | 19.0 | (144) | 8.6 | (6) | ||
| Exclusively | 9.0 | (93) | 5.3 | (8) | 16.3 | (123) | 7.1 | (5) | ||
| PMTCT of HIV | ||||||||||
| All | 36.1 | (375) | 76.2 | (115) | < .01 | 19.1 | (144) | 65.7 | (46) | < .01 |
| Associated with other indications | 20.4 | (212) | 24.5 | (37) | 13.5 | (102) | 35.7 | (25) | ||
| Exclusively | 15.7 | (163) | 51.7 | (78) | 5.6 | (42) | 30.0 | (21) | ||
a Failure to progress (16%), breech (10%), hypertension/preeclampsia (9%), failed induction (7%), postdates (6%), diabetes (5%), placenta previa (4%)
b Abnormal fetal heart rate (65%), intrauterine growth restriction (11%), meconium staining (9%), oligohydramnios (6%), multiple pregnancy (4%).
In women with VL <400 copies/mL (n = 4300), vaginal delivery increased with timely booking or HIV diagnosis (but not with GA at ARV initiation), CD4 count ≥200/μL, and type of maternity ( Table 2 ). Vaginal delivery was less frequent in earlier years, twins, VL ≥50 copies/mL, outside the Paris area, and moderate preterm delivery (32-36 weeks’ GA). All these variables, except time at booking, remained independently associated with vaginal delivery in multivariate analysis. Repeat cesareans were the main indication for elective CS (45.4%) and the exclusive one in 20.8%. Other obstetrical and/or maternal indications and fetal indications concerned 37.1% and 21.9%, respectively. Concern over the risk of MTCT was associated with these indications for nearly 20% of women, and was mentioned as the sole reason for 15.7% of elective CS ( Table 3 ). Nonelective CS were mostly for obstetrical reasons; PMTCT was mentioned for 19.1%, and as the sole reason for 5.6%.
Postpartum complications 2005 through 2010
Immediate postpartum complications and prolonged hospitalizations were more frequent following CS (elective or nonelective) than vaginal deliveries (6.5% vs 2.9% for complications, P < .01 and 15.6% vs 5.7% for prolonged hospitalization, P < .01) ( Table 4 ). Complications were mainly infections or hemorrhage. Maternal complications were the main reason for prolonged hospitalization. Complications and prolonged hospitalizations were more frequent in case of low maternal CD4 cell counts, whatever the mode of delivery ( Table 3 ).
| Variable | Total | According to mode of delivery | P value | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Vaginal delivery | CS nonelective | CS elective | |||||||
| n = 4001 | n = 1894 | n = 868 | n = 1239 | ||||||
| % | (n/N) | % | (n/N) | % | (n/N) | % | (n/N) | ||
| ≥1 complication a | 4.8 | (190/3977) | 2.9 | (54/1887) | 7.8 | (67/859) | 5.6 | (69/1231) | < .01 |
| Types of complication | |||||||||
| Hemorrhage requiring blood transfusion | 1.6 | (64) | 1.3 | (24) | 1.9 | (16) | 1.9 | (24) | |
| Infection(s) | 2.9 | (115) | 1.7 | (33) | 5.3 | (46) | 2.9 | (36) | |
| Endometritis | 0.3 | (11) | 0.2 | (3) | 0.6 | (5) | 0.2 | (3) | |
| Febrile urinary tract infection | 0.4 | (16) | 0.2 | (3) | 1.0 | (9) | 0.3 | (4) | |
| Abscess | 0.5 | (20) | 0.1 | (2) | 1.6 | (14) | 0.2 | (3) | |
| Pneumopathy | 0.1 | (5) | 0.1 | (1) | 0.2 | (2) | 0.2 | (2) | |
| Fever of undetermined origin b | 0.9 | (35) | 1.1 | (20) | 0.5 | (4) | 0.9 | (11) | |
| Other | 1.2 | (49) | 0.7 | (13) | 2.2 | (19) | 1.4 | (17) | |
| Deep vein thrombosis | 0.3 | (11) | 0.2 | (4) | 0.3 | (3) | 0.3 | (4) | |
| Anesthesiology complications | 0.1 | (2) | 0.0 | (0) | 0.1 | (1) | 0.1 | (1) | |
| Complications according to CD4 cells/μL | |||||||||
| <200 | 9.1 | (23/254) | 7.6 | (6/79) | 9.7 | (7/72) | 9.7 | (10/103) | .86 |
| 200-350 | 5.9 | (47/795) | 4.5 | (15/333) | 9.4 | (19/202) | 5.0 | (13/260) | .05 |
| 350-500 | 4.7 | (53/1132) | 1.8 | (9/498) | 8.3 | (21/254) | 6.1 | (23/380) | < .01 |
| ≥500 | 3.7 | (66/1775) | 2.5 | (24/969) | 5.8 | (19/327) | 4.8 | (23/479) | < .01 |
| Requiring postpartum surgery | 1.2 | (40/3281) | 0.8 | (13/1545) | 1.9 | (14/729) | 1.3 | (13/1007) | .09 |
| Hospitalization >7 d | 10.9 | (357/3264) | 5.7 | (87/1534) | 15.0 | (109/725) | 16.0 | (161/1005) | < .01 |
| Reasons for hospitalization | |||||||||
| Maternal complication | 2.2 | (71) | 0.7 | (11) | 3.9 | (28) | 3.2 | (32) | |
| Neonatal complication | 0.4 | (12) | 0.1 | (1) | 1.1 | (8) | 0.3 | (3) | |
| Psychosocial problems | 0.8 | (25) | 0.2 | (3) | 1.9 | (14) | 0.8 | (8) | |
| Other | 0.5 | (17) | 0.1 | (2) | 0.6 | (4) | 1.1 | (11) | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
