Abstract
BACKGROUND:
Despite some clinician advocacy for the use of gabapentin to treat neonatal irritability of presumed neurologic origin, the extent of gabapentin administration to hospitalized neonates is unknown. We aimed to identify trends in gabapentin utilization among infants hospitalized in neonatal intensive care units (NICUs) across the United States (US) and to evaluate the associations between clinical diagnoses and gabapentin treatment.
METHODS:
We analyzed NICU admitted neonates within the 2005–2016 Pediatric Health Information System to measure treatment timing, duration, and frequency. We used a modified Poisson regression with a robust between-cluster variance estimator to calculate a probability (adjusted relative risk) for gabapentin administration.
RESULTS:
Of 278,403 neonates, 374 were administered gabapentin treatment (0.13%). The median treatment duration was 16 days (25th-75th%: 8–40). Gabapentin use increased from 0% in 2005 to 0.39% in 2016. Treatment was prescribed to neonates at 31 of 48 studied hospitals; 73% of total treated infants localized to 5 NICUs. Term (0.16%) and ≤28-weeks gestation preterm infants (0.22%) were most likely to receive gabapentin. Varying by gestational age, a diagnosis of chromosomal abnormalities, severe bronchopulmonary dysplasia (BPD), hemorrhagic stroke, and neonatal abstinence syndrome (NAS) were associated with higher treatment with gabapentin. The majority (88.8%) of treated infants did not have a seizure diagnosis.
CONCLUSION:
Gabapentin use in US NICUs increased in recent years and varies markedly between institutions. Term infants, ≤28-weeks gestation preterm infants, and neonates with chronic genetic, neurologic, and gastrointestinal diagnoses were more likely to receive gabapentin.
Keywords: gabapentin, neonate, neonatal intensive care unit, off-label drug use, neurologic irritability
INTRODUCTION
Among neonates hospitalized in neonatal intensive care units (NICUs), a sometimes unclear distinction between pain and agitation can lead to the concept of irritability, a poorly defined, broad term describing subjective behavioral and physiological indicators such as persistent crying, tense posture and muscle tone, inconsolability, and vital sign changes.1–4 Nonverbal communication and a lack of well validated pain assessment scales create challenges for clinicians evaluating and treating perceived discomfort.1,2,5,6 In the NICU, various identifiable etiologies of neurological compromise include infections, ischemia, hemorrhage, thrombosis, and epileptogenic disorders. However, even in these infants presumed to experience chronic pain with agitation, the use of traditional pharmacologic management – including benzodiazepines and opioids – is varyingly effective.2–3 Adverse effects often occur, such as sedation, decreased gastrointestinal motility, and respiratory depression7,8 and the development of dependency and associated withdrawal symptoms after a few days of use.7,8 Consequently, alternative treatment options including gabapentin have been proposed as potentially less harmful treatment options.4,9,10
Gabapentin ([1-(aminomethyl)cyclohexaneaceticacid], Neurontin®11), a gamma-aminobutyric acid (GABA) analog,12 is approved by the Food and Drug Administration (FDA) to treat partial epilepsy in children as young as 3 years and post-herpetic neuralgia in adults.13,14 Gabapentin has and continues to be used in cases of seizures, pain, and irritability for neonates2–4 despite no formal FDA-approved indication for neonatal use. While studies are available examining adult use for approved indications and off-label use14–16, little research exists regarding gabapentin safety and effectiveness among a neonatal specific population.2–4, 15 In an era where clinicians are increasingly prescribing gabapentin despite this knowledge gap3, we aimed to examine trends in gabapentin administration among a national cohort of NICU-hospitalized neonates and to determine any associations between clinical diagnoses and increased probability of gabapentin treatment. Our objectives were to identify trends in gabapentin utilization among infants hospitalized in NICUs at United States (US) children’s hospitals and to evaluate the associations between clinical diagnoses and gabapentin treatment.
METHODS
Study Source
We conducted our analysis using de-identified data obtained from the Pediatric Health Information System (PHIS), a large administrative database (Children’s Hospital Association; Shawnee Mission, Kansas) containing demographic data, daily medication records, procedures, diagnoses, and billing information from 48 NICUs across the United States. This study was reviewed and approved by the Nationwide Children’s Institutional Review Board.
Inclusion Criteria and Variables
Our cohort included neonates discharged between January 2005 to June 2016 with a recorded gestational age (GA) at birth. A Clinical Transaction Classification (CTC) code was used to identify gabapentin administration (CTC code: 116047) and other medications (supplementary Table 1). Although we anticipate that neonatologists were the most common gabapentin prescribers within the NICU, and that consulting pediatric neurologists and developmental pediatricians may have influenced treatment decisions, we were unable to ascertain the exact specialty of the prescribing clinician. Clinical diagnoses included stages 2 and 3 of necrotizing enterocolitis (NEC) using x-ray documented signs of pneumatosis and perforation, chromosomal abnormalities, periventricular leukomalacia, seizures, congenital brain abnormalities, hypoxic ischemic encephalopathy (HIE), ischemic or thrombotic stroke, hemorrhagic stroke, grades 3 and 4 intraventricular hemorrhage (IVH), and Neonatal Abstinence Syndrome (NAS). Diagnoses and GA categories (22–24, 25–28, 29–32, 33–34, 35–36, and ≥37 weeks) were classified as binary variables using International Statistical Classification of Diseases and Related Health Problems, Ninth (ICD-9) and Tenth revision (ICD-10) diagnostic codes (supplementary Table 2). Preterm infants born at ≤32-weeks GA receiving oxygen were identified as having bronchopulmonary dysplasia (BPD) based on the degree of respiratory support at 36-weeks postmenstrual age (PMA): infants utilizing oxygen therapy were identified as having moderate BPD and infants using invasive or non-invasive positive pressure ventilation were identified as having severe BPD.17
Statistical Analysis
Treatment timing and duration of gabapentin administration was examined using univariate analysis. We estimated adjusted relative risks (adjusted risk ratios) (aRR) using a robust error variance Poisson regression, controlling for within-hospital clustering to measure any associations between various diagnoses and the probability of receiving gabapentin.18,19 The aRR is the multivariable adjusted ratio of the probability of gabapentin administration for patients with a measured risk factor (predictor variable) to the probability of gabapentin administration for patients without that same risk factor. Trends over time were evaluated with Cochran-Armitage tests. All analyses were conducted using Stata 15.0 (StataCorp, College Station, Texas). Since the purpose of our manuscript was to explore the prevalence of and factors associated with neonatal gabapentin utilization within specific patient subgroups, we chose not to control for multiple comparisons.20–21 This should be noted when interpreting our findings.
RESULTS
Patient Characteristics
Throughout the study period, 278,403 infants hospitalized in 48 PHIS-participating NICUs fit our inclusion criteria (female: n=122,384, 44%; ≥37 weeks GA: n=138,234, 50%) (Table 1). A total of 374 (0.13%) infants received gabapentin during their hospitalization in the NICU (female: n=153, 41%; ≥37 weeks GA: n=216, 58%). Of those treated with gabapentin, 12.0% (n=45) had severe BPD, 12% (n=45) were diagnosed with congenital brain abnormalities, 11.2% (n=42) with seizures, 10.7% (n=40) with chromosomal abnormalities, and 6.7% (n=25) with NAS. Within the treated cohort, 73 (19.5%) received gabapentin within the first 30 days of life and 262 (70.1%) still received gabapentin at discharge. A further breakdown of gabapentin administration and diagnoses by GA is available in Table 2.
Table 1:
Cohort demographic characteristics.
|
Full sample. N=278.403 n (%)* |
Gabapentin treated, N=374 n (%) |
|
|---|---|---|
| Birth gestation (weeks) | ||
| ≤28 | 32,092 (11.5) | 70 (18.7) |
| 29–32 | 31,546 (11.3) | 23 (6.2) |
| 33–36 | 76,531 (27.5) | 65 (17.4) |
| ≥37 | 138,234 (49.7) | 216 (57.8) |
| Sex | ||
| Female | 122,384 (44.0) | 153 (41.0) |
| Male | 155,869 (56.0) | 220 (59.0) |
| Diagnoses | ||
| NEC | 3,689 (1.3) | 16 (4.3) |
| Chromosomal abnormalities | 5,164 (1.9) | 40 (10.7) |
| Periventricular leukomalacia | 2,760 (1.0) | 14 (3.7) |
| Moderate BPD# | 7,276 (2.6) | 8 (2.1) |
| Severe BPD# | 6,524 (2.3) | 45 (12.0) |
| Seizure | 11,180 (4.0) | 42 (11.2) |
| Congenital brain abnormalities | 10,382 (3.7) | 45 (12.0) |
| HIE | 6,487 (2.3) | 21 (5.6) |
| Ischemic or thrombotic stroke | 1,334 (0.5) | 2 (0.5) |
| Hemorrhagic stroke | 682 (0.2) | 3 (0.8) |
| Intraventricular hemorrhage | 6,731 (2.4) | 14 (3.7) |
| NAS | 11,725 (4.2) | 25 (6.7) |
| Medications | ||
| Opioids | 109,617 (39.4) | 365 (97.6) |
| Benzodiazepine | 75,109 (27.0) | 343 (91.7) |
| Non-benzodiazepine antiepileptics | 22,434 (8.1) | 100 (26.7) |
| Dexmedetomidine | 6,583 (2.4) | 88 (23.5) |
| Propofol | 28,784 (10.3) | 155 (41.4) |
| Ever received gabapentin | 374 (0.13) | 374 (100) |
| Early use of gabapentin† | 73 (0.03) | 73 (19.5) |
| Discharged on gabapentin | 262 (0.09) | 262 (70.1) |
| Treatment duration (days)‡ | - | 16 (8–40) |
Abbreviations: NEC, necrotizing enterocolitis; BPD, bronchopulmonary dysplasia; HIE, hypoxic ischemic encephalopathy; NAS, neonatal abstinence syndrome.
Percentages were derived using N as the denominator;
BPD diagnoses restricted to infants born at ≤32 weeks of gestation;
Received within the first 30 days of life.;
X (X- X) = median (IQR).
Table 2:
Gabapentin treated cohort characteristics by gestational age categories.
| Gabapentin treated, N=374 | ||||
|---|---|---|---|---|
| Birth gestation (weeks) | ≤28, N= 70 n (%)* |
29–32, N=23 n (%) |
33–36, N= 65 n (%) |
≥37, N= 216 n (%) |
| Sex | ||||
| Female | 27 (38.6) | 5 (21.7) | 28 (43.1) | 93 (43.1) |
| Male | 43 (61.4) | 18 (78.3) | 37 (56.9) | 122 (56.7) |
| Diagnoses | ||||
| NEC | 11 (15.37) | 1 (4.4) | 1 (1.5) | 3 (1.4) |
| Chromosomal abnormalities | 1 (1.4) | 3 (13.0) | 11 (16.9) | 25 (11.6) |
| Periventricular leukomalacia | 8 (11.4) | 2 (8.7) | 1 (1.5) | 3 (1.4) |
| Moderate BPD# | 6 (8.6) | 2 (8.7) | - | - |
| Severe BPD# | 36 (51.4) | 9 (39.1) | - | - |
| Seizure | 10 (14.3) | - | 11 (16.9) | 21 (9.7) |
| Congenital brain abnormalities | 4 (5.7) | 5 (21.7) | 11 (16.9) | 25 (11.6) |
| HIE | - | 1 (4.4) | 8 (12.3) | 12 (5.6) |
| Ischemic or thrombotic stroke | - | - | - | 2 (0.9) |
| Hemorrhagic stroke | - | 2 (8.7) | - | 1 (0.5) |
| Intraventricular hemorrhage | 11 (15.7) | 1 (4.4) | 1 (1.5) | 1 (0.5) |
| NAS | 3 (4.3) | 4 (17.4) | 1 (1.5) | 17 (7.9) |
| Medications | ||||
| Opioids | 70 (100) | 21 (91.3) | 63 (96.9) | 211 (97.7) |
| Benzodiazepine | 66 (94.3) | 19 (82.6) | 60 (92.3) | 198 (91.7) |
| Non-benzodiazepine antiepileptics | 25 (35.7) | 7 (30.4) | 18 (27.7) | 50 (23.2) |
| Dexmedetomidine | 15 (21.4) | 9 (39.1) | 20 (30.8) | 44 (20.4) |
| Propofol | 25 (35.7) | 11 (47.8) | 40 (61.5) | 79 (36.6) |
| Early use of gabapentin† | - | 1 (4.4) | 9 (13.9) | 63 (29.2) |
| Discharged on gabapentin | 47 (67.1) | 19 (82.6) | 39 (60) | 157 (72.7) |
| Treatment duration (days)‡ | 31 (14–57) | 32 (12–61) | 22 (10–65) | 12 (7–27) |
Abbreviations: NEC, necrotizing enterocolitis; BPD, bronchopulmonary dysplasia; HIE, hypoxic ischemic encephalopathy; NAS, neonatal abstinence syndrome.
Percentages were derived using N as the denominator;
BPD diagnoses restricted to infants born at ≤32 weeks of gestation;
Received within the first 30 days of life.;
X (X- X) = median (IQR).
Treatment Timing and Duration
The median age at first administration of gabapentin was 66 postnatal days (25th-75th%: 37–125; Table 3). Those born ≥37 weeks GA had the earliest median age at first dose (47 days; 25th-75%: 30–78) and those ≤28 weeks had the latest (158 days, 25th-75th: 120–242; 32–74 days after 40 weeks’ post-term corrected age) (Table 3). The distribution of gabapentin administration by GA was bimodal. The age groups with the highest proportion of infants receiving gabapentin were ≤24 week preterm infants (0.31%) and ≥37-week term infants (0.16%) (Figure 1). Clinical diagnoses and sex had an additional impact on the timing of the first dose of gabapentin. Across most gestational age groups, male infants typically received gabapentin earlier than females (Table 3). Compared to those in the same GA stratum, the median age at first dose was earliest for infants diagnosed with NAS, HIE, chromosomal abnormalities, and IVH.
Table 3:
Median postnatal age at first gabapentin administration by gestational age.
| Median age at first dose* Postnatal days (25th-75th percentile) | |||||
|---|---|---|---|---|---|
| Full cohort | ≤28 weeks GA | 29–32 weeks GA | 33–36 weeks GA | ≥37 weeks GA | |
| All treated patients | 66 (37–125) | 158 (120–242) | 101 (78–155) | 74 (47–121) | 47 (30–78) |
| Diagnoses | |||||
| NEC | 144 (110–228) | 170 (120–235) | 278 (278–278) | 115 (115–115) | 96 (69–116) |
| Chromosomal abnormalities | 64 (42–92) | 115 (115–115) | 100 (71–101) | 92 (47–171) | 53 (41–77) |
| Periventricular leukomalacia | 107 (90–148) | 146 (105–241) | 111 (90–131) | 106 (106–106) | 59 (27–96) |
| Moderate BPD# | 140 (113–155) | 140 (124–154) | 128 (101–155) | - | - |
| Severe BPD# | 144 (107–182) | 157 (121–202) | 100 (74–144) | - | - |
| Seizures | 71 (32–141) | 172 (148–253) | - | 81 (35–125) | 38 (29–70) |
| Congenital brain abnormalities | 75 (35–125) | 137 (114–335) | 134 (101–144) | 81 (56–125) | 41 (32–76) |
| HIE | 30 (13–49) | - | 27 (27–27) | 51 (23–60) | 25 (13–39) |
| Ischemic or thrombotic stroke | 42 (39–45) | - | - | - | 42 (39–45) |
| Hemorrhagic stroke | 145 (100–155) | - | 128 (100–155) | - | 145 (145–145) |
| Intraventricular hemorrhage | 141 (91–170) | 166 (127–235) | 134 (134–134) | 59 (59–59) | 22 (22–22) |
| NAS | 32 (26–71) | 103 (80–172) | 62 (43–175) | 20 (20–20) | 29 (22–38) |
| Sex | |||||
| Female | 71 (42–120) | 170 (120–260) | 155 (84–157) | 66 (43–103) | 55 (38–77) |
| Male | 61 (33–129) | 148 (117–231) | 101 (78–134) | 81 (49–139) | 38 (27–60) |
Abbreviations: NEC, necrotizing enterocolitis; BPD, bronchopulmonary dysplasia; HIE, hypoxic ischemic encephalopathy; NAS, neonatal abstinence syndrome.
See Table 2 for patient n included in each calculation.;
BPD diagnoses restricted to infants born at ≤32 weeks of gestation.
Figure 1:
Gabapentin utilization by gestational age at birth.
Overall, the median treatment duration for all treated infants was 16 days (25th-75th%: 8–36; Table 1a). Infants ≤32-weeks had a median duration of 32 days (≤28 weeks: 31; 25th-75th%: 14–57) (29–32 weeks: 32; 25th-75th%: 12–61) and those born at >32-weeks had a median duration of 14 days (33–36 weeks: 22; 25th-75th%: 10–65) (≥37 weeks: 12; 25th-75th%:7–27).
Clinical Predictors of Gabapentin Administration
After accounting for gestational age, sex, and other diagnoses (Table 4), the largest measured predictors of ever receiving gabapentin among the entire cohort were diagnoses of hemorrhagic stroke, NAS, and severe BPD for infants, differing by GA (Table 4). Neonates with severe BPD ranged from 5.20 (95% confidence interval (CI): 3.19, 8.49) to 7.43 (95%: 3.23, 17.06) times greater probability of being treated with gabapentin compared to neonates not diagnosed with severe BPD within the same GA categories. Conversely, infants with a diagnosis of intraventricular hemorrhage born at ≤28 weeks GA had a 0.61 (95% CI: 0.40, 0.92) times lower likelihood of receiving gabapentin than those within the same GA category that did not have an intraventricular hemorrhage. A lower probability was also seen among females born between 29–32 weeks GA (aRR: 0.34, 95%: 0.15, 0.76). The magnitude of these associations stratified by GA can further be seen in Table 4.
Table 4:
Multivariable adjusted probability of receiving gabapentin.
| aRR* (95% CI) | ||||
|---|---|---|---|---|
| Birth gestation | ≤28 weeks | 29–32 weeks | 33–36 weeks | ≥37 weeks |
| Diagnoses | ||||
| NEC | 1.96 [1.13, 3.43] | 0.81 [0.08, 8.25] | 1.63 [0.24, 11.07] | 3.74 [1.48, 9.46] |
| Chromosomal abnormalities | 1.80 [0.39, 8.37] | 3.70 [1.06, 12.93] | 6.42 [3.21, 12.83] | 4.77 [2.50, 9.11] |
| Periventricular leukomalacia | 1.70 [0.85, 3.41] | 1.89 [0.34, 10.45] | 1.68 [0.20, 14.05] | 4.16 [1.14, 15.23] |
| Moderate BPD# | 0.90 [0.17, 4.91] | 1.81 [0.42, 7.77] | - | - |
| Severe BPD# | 5.20 [3.18, 8.48] | 7.43 [3.23, 17.06] | - | - |
| Seizure | 3.61 [1.42, 9.21] | - | 3.98 [1.90, 8.36] | 1.48 [0.85, 2.57] |
| Congenital brain abnormalities | 0.73 [0.31, 1.69] | 4.75 [1.62, 13.94] | 2.91 [1.03, 8.23] | 2.37 [0.93, 6.04] |
| HIE | - | 4.07 [1.03, 16.08] | 4.03 [1.79, 9.07] | 1.42 [0.48, 4.21] |
| Ischemic or thrombotic stroke | - | - | - | 0.95 [0.33, 2.79] |
| Hemorrhagic stroke | - | 15.52 [1.23, 195.14] | - | 1.58 [0.18, 13.73] |
| Intraventricular hemorrhage | 0.61 [0.40, 0.92] | 0.68 [0.05, 9.09] | 0.97 [0.14, 6.95] | 0.95 [0.11, 8.02] |
| NAS | 1.96 [0.98, 3.94] | 8.53 [2.97, 24.50] | 0.45 [0.06, 3.60] | 1.63 [0.55, 1.13] |
| Sex | ||||
| Female | 0.81 [0.49, 1.35] | 0.34 [0.15, 0.76] | 0.91 [0.55, 1.50] | 0.95 [0.79, 1.13] |
| ICC† =0.09 [0.06, 0.13] | ||||
Abbreviations: NEC, necrotizing enterocolitis; BPD, bronchopulmonary dysplasia; HIE, hypoxic ischemic encephalopathy; NAS, neonatal abstinence syndrome; ICC, intraclass coefficient.
See Table 2 for patient n included in each calculation;
BPD diagnoses restricted to infants born at ≤32 weeks of gestation;
The ICC, the proportion of total variance in gabapentin use due to variation between hospitals22, indicated that clustering by hospital was a significant component of the overall variation in the frequency of gabapentin treatment.
Hospital Utilization of Gabapentin
The proportion of neonates that received gabapentin increased rapidly throughout the study period, from no use in 2005 to 0.07% in 2010, followed by a steep increase to 0.39% in 2016 (Figure 2). When examining interhospital variation in gabapentin administration, gabapentin was prescribed in 31 of 48 studied hospitals and 72.5% of infants ever treated with gabapentin were hospitalized at 5 centers (Figure 3).
Figure 2:
Gabapentin utilization by discharge year.
Figure 3:
Interhospital variation of gabapentin treatment among NICU admitted neonates.
Other Medications
Over 90% of gabapentin treated neonates received benzodiazepines or opioids during at least one service day of their NICU time (Figure 4). A total of 370 (98.9%) infants were prescribed other opioid or anxiolytic sedative medications in conjunction with gabapentin and 217 (58%) received these other medications concurrently at the time of gabapentin initiation.
Figure 4:
Percent of gabapentin treated infants ever treated with or concurrently taking other medications.
DISCUSSION
We observed a significant, rapid increase in national neonatal gabapentin administration throughout our 11-year study period despite a lack of high quality safety and efficacy evidence.2–4,15,23 Clinical diagnoses were associated with both increased and decreased probabilities of receiving gabapentin. We unexpectedly observed a high likelihood of gabapentin treatment for infants with severe BPD. Currently, there are no proven mechanisms or disorders supporting a neuropathic origin of pain in BPD. Furthermore, 58% of our treated cohort received opioid or sedative medications on the initial day of gabapentin administration. Our findings of large institutional variability in gabapentin prescription rates in a setting of limited effectiveness or safety data may be reflective of subjectivity in both pain assessment measures and treatment decisions, resulting in inconsistent management.6 Gabapentin is used as an analgesic for cases of chronic pain24, but the prolonged effects of both gabapentin and pain on the early, active processes of neonatal neurodevelopment are unknown. Thus, physicians must weigh symptom alleviation with long-term effects of the underlying diagnosis.
Metabolically, gabapentin does not bind to and affect GABAA receptor function despite a structural similarity to GABA.25,26 GABA acts as an inhibitory neurotransmitter in adults but contrastingly creates an excitatory effect in the neonatal brain.27–29 This resulting neuronal membrane depolarization is a hypothesized critical component of early cortical development.27–29 Gabapentin has been shown in rodent models and human-based in vitro systems to selectively inhibit the alpha-2 delta-1 (α2δ−1) subunit of voltage-gated calcium channels encoded by the CACNA2D1 gene,30 thereby alleviating neuropathic pain. Further investigation is warranted to determine whether treatment in neonates causes increased GABA levels or α2δ−1 inhibition.
Emerging concern exists regarding possible neonatal gabapentin withdrawal31 after several studies contend a possibility of fetal retention following maternal exposure.32–34 Huybrechts et al.35 reported in utero co-exposure to opioids and gabapentin was associated with both an increased risk of neonatal withdrawal and more severe withdrawal symptoms compared to opioid exposure alone; a worrisome finding given 58% of our treated cohort were concomitantly prescribed opioid or sedative medications and gabapentin. Moreover, two case reports31,34 featuring neonates withdrawing from in utero gabapentin exposure reported symptom improvement following a reintroduction to gabapentin. In an era of rising co-prescription rates of gabapentin, opioids, and/or benzodiazepines36 and gabapentin misuse among adults,13,37 the weighing of perceived benefits against real dependency risks is imperative.2–4, 15
Few published studies indicate that gabapentin is an effective treatment for refractory pain and irritability3–4. Edwards et al.3 reported improved feeding tolerance and decreased irritability among a sample of 11 gabapentin treated neonates with “visceral hyperalgesia.” However, no quantitative or physiologic measurement was used to establish the initial diagnosis prompting drug usage. Adverse events were recorded in 5 neonates; three experienced tachycardia, emesis, and irritability as a result of abrupt discontinuation of gabapentin and two experienced bradycardia resolved with a lower dose of gabapentin. Most recently, Sacha et al.5 published similar findings for 22 NICU admitted neonates given gabapentin therapy for pain and agitation. Neonatal Pain, Agitation and Sedation Scale (N-PASS) scores1 improved and administration of other analgesic and sedative medications decreased. One patient experienced nystagmus. Despite these studies’ small samples, they reflect gabapentin’s seeming ability to improve pain-attributed symptoms among a neonatal population.
Our study is primarily strengthened by including a large sample size that enabled us to measure gabapentin utilization and examine associations between clinical diagnoses and gabapentin treatment, aspects that have not been evaluated to this scale in the few similar studies available. Conversely, the use of a large administrative database is inherently prone to errors of coding, misclassification, and underreporting but the effect may be minimized by internal PHIS database quality checks.38 We were unable to evaluate non-recorded clinical signs that may have provoked gabapentin treatment, analyze follow-up data post-discharge, and examine for adverse effects post-treatment given that we relied on PHIS discharge diagnoses and not direct medical records. Moreover, gabapentin prescription was subjective based on physician and institutional preferences. This may impact the ability to generalize our results. However, it importantly highlights the lack of a national or global consensus on neonatal gabapentin treatment, presumably due to a lack of evidence on safety or effectiveness.
CONCLUSION
Our data demonstrates an increasing frequency of gabapentin use in U.S. NICUs. The majority of treated infants were not diagnosed with seizures and based on their concurrent diagnoses and previous reports,2–5 it is possible that they were instead treated for neonatal irritability and agitation of unproven origin. Neonates diagnosed with chromosomal abnormalities, necrotizing enterocolitis, periventricular leukomalacia, congenital brain abnormalities, BPD, and seizures had higher likelihood of receiving gabapentin. Prescription rates varied greatly between institutions, highlighting a lack of cohesive guidelines. Given concerns for effectiveness, safety, long-term effects, and potential dependence, further rigorous prospective investigations within the neonatal population are urgently needed.
Supplementary Material
ACKNOWLEDGEMENTS
This work was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health [K08HL121182 to Dr.Slaughter]. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Declarations of interest: none.
REFERENCES
- [1].Hummel P, Puchalski M, Creech SD, Weiss MG. Clinical reliability and validity of the N-PASS: neonatal pain, agitation and sedation scale with prolonged pain. Journal of Perinatology. 2008;28(1):55–60. doi: 10.1038/sj.jp.7211861. [DOI] [PubMed] [Google Scholar]
- [2].Haney AL, Garner SS, Cox TH. Gabapentin therapy for pain and irritability in a neurologically impaired infant. Pharmacotherapy. 2009;29(8):997–1001. [DOI] [PubMed] [Google Scholar]
- [3].Edwards L, DeMeo S, Hornik CD, et al. Gabapentin use in the neonatal intensive care unit. J Pediatr. 2015;169:310–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Sacha GL, Foreman MG, Kyllonen K, Rodriguez RJ. The use of gabapentin for pain and agitation in neonates and infants in a neonatal ICU. J Pediatr Pharmacol Ther. 2017;22(3):207–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Hauer JM, Wical BS, Charnas L. Gabapentin successfully manages chronic unexplained irritability in children with severe neurologic impairment. Pediatrics. 2007;119(2):e519–e522. doi: 10.1542/peds.2006-1609. [DOI] [PubMed] [Google Scholar]
- [6].Hall RW, Anand KJ. Pain management in newborns. Clin Perinatol. 2014;41(4):895–924. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Walter-Nicolet E, Annequin D, Biran V, Mitanchez D, Tourniaire B. Pain management in newborns: from prevention to treatment. Pediatric Drugs. 2010;12(6):353–365. [DOI] [PubMed] [Google Scholar]
- [8].Hudak ML, Tan RC. Neonatal drug withdrawal. Pediatrics. 2012;129(2):e540–e560. doi: 10.1542/peds.2011-3212. [DOI] [PubMed] [Google Scholar]
- [9].Alles SRA, Smith PA. Etiology and pharmacology of neuropathic pain. Pharmalogical Reviews. 2018;70(2):315–347. [DOI] [PubMed] [Google Scholar]
- [10].Hauer J, Mackey D. Treatment with gabapentin associated with resolution of apnea in two infants with neurologic impairment. Journal of Palliative Medicine. 2013;16(4):455–458. doi: 10.1089/jpm.2012.0103. [DOI] [PubMed] [Google Scholar]
- [11].Haig G, Bockbrader H, Wesche D, et al. Single-dose gabapentin pharmacokinetics and safety in healthy infants and children. J Clin Pharmacol. 2001;41(5):507–514. [DOI] [PubMed] [Google Scholar]
- [12].Behm MO, Kearns GL. Treatment of pain with gabapentin in a neonate. Pediatrics. 2001;108(2):482–484. doi: 10.1542/peds.108.2.482. [DOI] [PubMed] [Google Scholar]
- [13].NEURONTIN®. Prescribing Information and Medication Guide 2017. Pfizer, Inc; New York,NY:https://www.accessdata.fda.gov/drugsatfda_docs/label/2017/020235s064_020882s047_021129s046lbl.pdf. [Google Scholar]
- [14].Wallach JD, Ross JS. Gabapentin approvals, off-label use, and lessons for postmarketing evaluation efforts. JAMA: Journal of the American Medical Association. 2018;319(8):776–778. doi: 10.1001/jama.2017.21897. [DOI] [PubMed] [Google Scholar]
- [15].Ouellet D, Bockbrader HN, Wesche DL, Shapiro DY, Garofalo E. Population pharmacokinetics of gabapentin in infants and children. Epilepsy Res. 2001;47(3):229–41. [DOI] [PubMed] [Google Scholar]
- [16].Moore A, Derry S, Wiffen P. Gabapentin for chronic neuropathic pain. JAMA: Journal of the American Medical Association. 2018;319(8):818–819. doi: 10.1001/jama.2017.21547. [DOI] [PubMed] [Google Scholar]
- [17].Ehrenkranz RA, Walsh MC, Vohr BR, et al. Validation of the national institutes of health consensus definition of bronchopulmonary dysplasia. Pediatrics. 2005;116(6):1353–1360. doi: 10.1542/peds.2005-0249. [DOI] [PubMed] [Google Scholar]
- [18].Zou G, Donner A. Extension of the modified Poisson regression model to prospective studies with correlated binary data. Statistical Methods in Medical Research. 2013;22(6):661–670. doi: 10.1177/0962280211427759. [DOI] [PubMed] [Google Scholar]
- [19].Williams RL. A note on robust variance estimation for cluster-correlated data. Biometrics. 2000; 56:645–6. [DOI] [PubMed] [Google Scholar]
- [20].Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology. 1990;1(1):43–46. [PubMed] [Google Scholar]
- [21].Rothman KJ. Six persistent research misconceptions. J Gen Intern Med. 2014;29(7):1060–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Killip S, Mahfoud Z, Pearce K. What is an intracluster correlation coefficient? Crucial concepts for primary care researchers. Annals of Family Medicine. 2004;2(3):204–208. doi: 10.1370/afm.141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Peckham AM, Evoy KE, Ochs L, Covvey JR. Gabapentin for off-label use: evidence-based or cause for concern?. Subst Abuse. 2018;12:1178221818801311 Published 2018 Sep 23. doi: 10.1177/1178221818801311 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Carter BS; Brunkhorst J Neonatal pain management. Semin. Perinatol 2017. 2017;41:111–116. doi: 10.1053/j.semperi.2016.11.001 [DOI] [PubMed] [Google Scholar]
- [25].Sills GJ. The mechanisms of action of gabapentin and pregabalin. Current Opinion in Pharmacology. 2006;6(1):108–113. doi: 10.1016/j.coph.2005.11.003. [DOI] [PubMed] [Google Scholar]
- [26].Patel R, Dickenson AH. Mechanisms of the gabapentinoids and α 2 δ−1 calcium channel subunit in neuropathic pain. Pharmacol Res Perspect. 2016;4(2):e00205. doi: 10.1002/prp2.205 [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Chapman KE, Raol YH, Brooks-Kayal A. Neonatal seizures: controversies and challenges in translating new therapies from the lab to the isolette. Eur J Neurosci. 2012;35(12):1857–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Leinekugel X, Medina I, Khaliliv I, Ben-Ari Y, Khazipov R. Ca2+ oscillations mediated by the synergistic excitatory actions of GABAA and NMDA receptors in the neonatal hippocampus. Neuron. 1997;18:243–255. doi: 10.1016/s0896-6273(00)80265-2 [DOI] [PubMed] [Google Scholar]
- [29].Donovan MD, Boylan GB, Murray DM, Cryan JF, Griffin BT. Treating disorders of the neonatal central nervous system: pharmacokinetic and pharmacodynamic considerations with a focus on antiepileptics. British Journal of Clinical Pharmacology. 2016;81(1):62–77. doi: 10.1111/bcp.12753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [30].Gee N, Brown J, Dissanayake V, Offord J, Thurlow R, Woodruff G. The novel anticonvulsant drug, gabapentin (Neurontin), binds to the alpha2delta subunit of a calcium channel. The Journal Of Biological Chemistry. 1996;271(10):5768–5776. [DOI] [PubMed] [Google Scholar]
- [31].Carrasco M, Rao S, Bearer C, Sundararajan S. Neonatal gabapentin withdrawal syndrome. Pediatric Neurology. 2015;53(5):445–447. [DOI] [PubMed] [Google Scholar]
- [32].Loudin S, Murray S, Prunty L, Davies T, Evans J, Werthammer J. An atypical withdrawal syndrome in neonates prenatally exposed to gabapentin and opioids. Journal Of Pediatrics. 2017;181:286–288. [DOI] [PubMed] [Google Scholar]
- [33].Öhman I, Vitols S, Tomson T. Pharmacokinetics of gabapentin during delivery, in the neonatal period, and lactation: does a fetal accumulation occur during pregnancy? Epilepsia. October 2005;46(10):1621–1624. [DOI] [PubMed] [Google Scholar]
- [34].Brzenski A, Greenberg, M. Neonatal abstinence syndrome due to in-utero exposure to gabapentin: a case report. International Journal of Medical and Pharmaceutical Case Reports. 2015; 3(5): 116–120. [Google Scholar]
- [35].Huybrechts K, Bateman B, Stover M, et al. Risk of neonatal drug withdrawal after intrauterine co-exposure to opioids and psychotropic medications: cohort study. BMJ (Clinical Research Ed.). 2017;358:j3326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [36].Montastruc F, Loo SY, Renoux C. Trends in first gabapentin and pregabalin prescriptions in primary care in the United Kingdom, 1993–2017. JAMA. 2018;320(20):2149–2151. doi: 10.1001/jama.2018.12358. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [37].Evoy K, Morrison M, Saklad S. Abuse and misuse of pregabalin and gabapentin. Drugs. 2017;77(4):403–426. [DOI] [PubMed] [Google Scholar]
- [38].Slaughter JL, Stenger MR, Reagan PB, Jadcherla SR. Neonatal H2-Receptor Antagonist and Proton Pump Inhibitor Treatment at US Children’s Hospitals. The Journal of pediatrics. 2016;174:63–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.