TABLE 2.
In-Hospital Mortality and Morbidities Among Infants Born at 22 to 28 Weeks’ GA by Maternal IDDM and Timing of Maternal Insulin Use
| Outcome, n (%)a | IBP, n = 312 | IDP, n = 195 | No IDDM, n = 10 245 | Adjusted RR (95% CI) for Outcomeb | ||
|---|---|---|---|---|---|---|
| IBP vs IDP | IBP vs No IDDM | IDP vs No IDDM | ||||
| Died before discharge | 85 (27) | 35 (18) | 2727 (27) | 1.33 (1.00–1.79) | 1.17 (0.99–1.38) | 0.87 (0.68–1.12) |
| Died ≤12 h | 26 (8) | 13 (7) | 1128 (11) | 1.15 (0.74–1.79) | 1.08 (0.79–1.48) | 0.94 (0.68–1.30) |
| Infants who survived >12 h | n = 286 | n = 182 | n = 9117 | |||
| RDS | 282 (99) | 182 (100) | 8934 (98) | — | 1.01 (1.00–1.03) | — |
| PDA | 130 (46) | 73 (40) | 4198 (46) | 1.20 (0.98–1.47) | 1.07 (0.94–1.21) | 0.89 (0.75–1.04) |
| NEC | 44 (15) | 16 (9) | 977 (11) | 1.67 (0.96–2.91) | 1.55 (1.17–2.05)c | 0.93 (0.57–1.52) |
| EOS | 6 (2) | 4 (2) | 187 (2) | 0.89 (0.25–3.10) | 1.09 (0.49–2.45) | 1.23 (0.47–3.24) |
| Infants in hospital >3 d | n = 275 | n = 179 | n = 8800 | |||
| LOS | 95 (35) | 36 (20) | 2458 (28) | 1.34 (0.97–1.85) | 1.26 (1.07–1.48)c | 0.94 (0.71–1.24) |
| Infants with cranial sonogram within 28 dd | n = 277 | n = 178 | n = 8892 | |||
| Severe IVH | 43 (16) | 24 (13) | 1382 (16) | 0.91 (0.58–1.44) | 1.02 (0.78–1.33) | 1.11 (0.77–1.61) |
| Infants with cranial sonogram within 28 d and/or closest to 36 wk PMA and after 28 de | n = 279 | n = 179 | n = 8914 | |||
| PVL | 8 (3) | 8 (4) | 410 (5) | 0.56 (0.21–1.47) | 0.64 (0.32–1.28) | 1.14 (0.57–2.27) |
| Infants with sufficient information to include in analysisf | n = 277 | n = 178 | n = 8886 | |||
| Severe IVH or PVL | 46 (17) | 24 (13) | 1550 (17) | 1.01 (0.65–1.57) | 0.97 (0.75–1.26) | 0.96 (0.67–1.39) |
| Infants in hospital at 28 d with ROP examinationg | n = 230 | n = 160 | n = 7569 | |||
| ROP | 130 (57) | 78 (49) | 4345 (57) | 1.23 (1.04–1.45)c | 1.09 (0.99–1.20) | 0.88 (0.77–1.01) |
| ROP stage ≥3 | 34 (15) | 16 (10) | 1018 (13) | 1.32 (0.78–2.24) | 1.27 (0.94–1.71) | 0.96 (0.62–1.49) |
| Infants alive at 36 wk PMA evaluated for BPDh | n = 233 | n = 164 | n = 7637 | |||
| BPD | 84 (36) | 58 (35) | 3355 (44) | 1.04 (0.81–1.33) | 0.93 (0.79–1.09) | 0.90 (0.74–1.08) |
Among survivors >12 h, information was missing for RDS, 1 infant; PDA, 10 infants; NEC, 2 infants; EOS, 3 infants. There were 1496 infants who died within 3 d of birth and 2 infants who stayed in the birth hospital <3 d (both were transferred) who were excluded from the number of infants in the hospital >3 d. LOS was missing for 1 infant who survived >3 d. The numbers of infants with missing values of other outcomes are noted in the following footnotes.
RRs and CIs from modified Poisson regression models fit to each outcome that in addition to the maternal insulin timing indicator (IBP, IDP, no IDDM) included study center, GA (categorical), male gender, antenatal steroid use, maternal race/ethnicity (black, white, Hispanic, other), and maternal age. For the outcome RDS, RRs involving infants whose mothers started insulin during pregnancy were not estimated because all infants in this group had RDS.
RRs significantly different from 1.0.
Severe IVH was defined as grade 3 or 4 and was diagnosed on the basis of the cranial sonogram taken within 28 d of birth with the most severe findings. Of 9585 infants who survived >12 h, 9347 (97.5%) were evaluated by cranial sonogram. Of those who had a cranial sonogram, IVH was missing for 4 infants.
PVL was determined on the basis of the cranial sonogram taken within 28 d of birth with the most severe findings and/or a cranial imaging study taken closest to 36 wk PMA and after 28 d of birth. Of the 9585 infants who survived >12 h, 9372 (97.8%) were evaluated for PVL. Among those with cranial imaging performed, PVL was missing for 4 infants.
Presence of IVH and/or PVL was determined for infants with nonmissing IVH and PVL outcomes, except that a diagnosis of either condition was sufficient to set the outcome.
ROP was defined for infants still hospitalized at 28 d who had a ROP examination. By 28 d, 2331 infants had died, 5 had been discharged, and status was unknown for 10. Of the 8406 infants still in the hospital, 7959 (94.7%) had a ROP examination. Among those who had an examination, ROP was missing for 1 infant.
BPD was defined for infants born <36 wk GA as the need for supplemental oxygen use at 36 wk PMA. For infants discharged or transferred before 36 wk PMA, BPD was defined based on oxygen use at 36 wk if known or oxygen use at the time of discharge or transfer. BPD could not be evaluated for 87 (1%) of the 8121 infants alive at 36 wk PMA. Of the 8034 infants evaluated, 82% were still in the hospital at 36 wk PMA, 12% had been discharged from the hospital, and 5% had been transferred to another hospital.