SARS-CoV-2 testing and clinical outcomes in a Texas tertiary care center labor and delivery unit

Abstract

Our hospital adopted universal severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) testing for labor and delivery patients in April 2020. The primary aim of this study was to determine the number of subjects from April 1, 2020, to July 31, 2020, who had laboratory-confirmed SARS-CoV-2 infection, and the secondary aims were to report demographic and clinical data for subjects with and without SARS-CoV-2 infection and the time from SARS-CoV-2 test collection to result for tests administered in the hospital. A total of 898 subjects had either vaginal or cesarean deliveries with a gestational age of >20 weeks during the study period. Of this group, 746 subjects underwent SARS-CoV-2 testing, and 16 had a positive test result. Four of the 16 subjects with laboratory-confirmed SARS-CoV-2 infection had documented symptoms at the time of admission. The difference in cohort size was too large to make a meaningful statistical comparison in demographic and clinical data between those with positive vs negative SARS-CoV-2 test results. The median time from SARS-CoV-2 test collection to result decreased from 239 minutes in April 2020 to 119 minutes in July 2020. Universal SARS-CoV-2 testing revealed a 2.1% positivity rate during our study period.

Keywords:

• Anesthesia
• COVID-19
• delivery
• obstetric

Sutton and colleagues reported that, following universal testing of subjects at their hospital from late March to early April 2020, 13.5% of patients presenting to the labor and delivery unit had laboratory-confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection but were asymptomatic.¹ This finding prompted our institution to adopt a universal testing strategy for the labor and delivery unit that was fully implemented in mid April 2020. This study reports on SARS-CoV-2 testing and clinical outcomes for subjects who delivered with a gestational age >20 weeks at Baylor Scott & White Medical Center – Temple from April 1, 2020, to July 31, 2020.

METHODS

After approval and waiver of informed consent by the Baylor Scott & White Research Institute institutional review board, the hospital delivery log was used to identify subjects who delivered during the study period. Demographic and clinical data were extracted by hand from the electronic medical record for analysis. This was a retrospective observational study, and no power analysis was conducted a priori. Median and interquartile ranges were calculated using Excel.

RESULTS

A total of 898 subjects met inclusion criteria, and 746 subjects had laboratory results for SARS-CoV-2 infection. Sixteen subjects had laboratory-confirmed SARS-CoV-2 infection, four of whom had documented symptoms at the time of admission. Of the 108 subjects with a pending SARS-CoV-2 test at the time of delivery, 4 and 104 subjects were subsequently confirmed to have positive and negative SARS-CoV-2 infection, respectively. A total of 429 subjects had SARS-CoV-2 testing performed during the hospital admission, and the median (interquartile) times from test collection to result were 239 minutes (183–552) in April, 195 minutes (164–289) in May, 189 minutes (131–384) in June, and 119 minutes (88–154) in July.

Testing platforms included Abbott ID Now (Abbott, Abbott Park, IL), 2 subjects, with a median time from test collection to results of 36 minutes (interquartile range, 36–36); Aries (Luminex, Austin, TX), 183 subjects, 188 minutes (165–237); GenXpert (Cepheid, Sunnydale, CA), 175 subjects, 106 minutes (85–148); Hologic Panther (Hologic, Marlborough, MA), 46 subjects, 520 minutes (418–832); Lyra (Quidel, San Diego, CA), 4 subjects, 790 minutes (509–1170); MagPix (Luminex, Austin, TX), 15 subjects, 1168 minutes (557–1657); and NxTag (Luminex, Austin, TX), 4 subjects, 460 minutes (235–782). Table 1 presents demographic and clinical data for subjects with SARS-CoV-2 test results.

Table 1. Demographic and clinical data for subjects in labor and delivery with a SARS-CoV-2 test result, April 1, 2020, to July 31, 2020

	Positive (N = 16)	Negative (N = 730)
Race
Not Hispanic	12	518
Hispanic	4	209
Patient declined	0	1
Unknown	0	2
Median age, years (IQR)	28 (23–32)	28 (24–32)
Median body mass index, kg/m^2 (IQR)^a	30.5 (28.0-37.9)	33.6 (29.0–38.5)
Median gravidity (IQR)	3 (1–3)	2 (1–3)
Median parity (IQR)	0 (0–2)	1 (0–2)
Multiple gestation	2	16
Prior cesarean delivery	2	147
Vaginal delivery	5	498
Spontaneous	5	475
Forceps assisted	0	15
Vacuum assisted	0	8
Cesarean delivery	11	232
Emergent	3	10
Regional anesthesia	8	201
General anesthesia	3	31
Median birthweight, g (IQR)^b	2830 (1720–3275)	3248 (2871–3584)
Median 1-min APGAR score (IQR)	8 (2–8)	8 (8–8)
Median 5-min APGAR score (IQR)	9 (7–9)	9 (9–9)
NICU involvement in resuscitation of neonate^c	13	258
Postoperative disposition of mother
Labor and delivery unit or monitored floor	15	724
Intensive care unit	1	6
Postoperative disposition of neonate
Labor and delivery unit	5	576
Neonatal intensive care unit	9	150
Neonatal demise	2	2
Intrauterine fetal demise	0	2

^a Available for 15 subjects in positive group and 597 subjects in negative group.

^b Available for 17 subjects in positive group and 742 subjects in negative group.

^c Available for 16 subjects in positive group and 730 subjects in negative group.

IQR indicates interquartile range; NICU, neonatal intensive care unit.

DISCUSSION

Our overall test positivity rate was 2.1% during the study period, and three-quarters of subjects with laboratory-confirmed SARS-CoV-2 infection did not have documented symptoms at the time of admission. The test positivity rate for SARS-CoV-2 in the county was not available from the Bell County Health Department, but new cases in the county (population 362,924) increased from 121 in April to 2212 in July.²

Collier and colleagues attempted to validate a point-of-care nucleic acid test for SARS-CoV-2 testing and found that the median time to test result was 2.6 hours compared with 26.4 hours for their standard laboratory method.³ We found that our median time for SARS-CoV-2 testing using nucleic acid amplification techniques decreased from 239 minutes in the month of April to 119 minutes in the month of July, which correlates to the approximately 156 minutes reported for the point-of-care test.

Pineles and colleagues reported that Hispanic subjects accounted for 73% and 55% of those who tested positive and negative for SARS-CoV-2, respectively, at a community hospital in Houston, Texas, suggesting that Hispanic patients were disproportionately represented in the positive group.⁴ In our study, Hispanic subjects accounted for 25% and 29% of subjects who tested positive and negative for SARS-CoV-2, respectively.

Farghaly and colleagues found that neonates born to mothers with laboratory-confirmed SARS-CoV-2 infection were more likely to have a cesarean delivery, lower APGAR scores, and a lower birthweight compared to neonates born to mothers without SARS-CoV-2 infection.⁵ In our study, the birthweight of neonates born to mothers with laboratory-confirmed SARS-CoV-2 infection was 2.83 kg, and approximately 70% of these neonates were delivered via cesarean section.

Among the limitations of our study, the SARS-CoV-2 status was unknown for 152 subjects; a meaningful statistical comparison of demographic and clinical data could not be made between subjects with laboratory-confirmed positive and negative SARS-CoV-2 infection due to the disparity in cohort size; and our hospital system has a practice of only asking patients whether their ethnicity is Hispanic or not. A strength of our study was that the official hospital labor and delivery log was used to capture every delivery during the study period, which ensured complete representation of all births.

ACKNOWLEDGMENTS

The authors acknowledge Michelle Hicks, BSN, for her assistance in gathering data from the Baylor Scott & White Medical Center – Temple delivery log.