Pyloric Stenosis in Premature Twins: A Case Report

Summary.

• Pyloric stenosis in premature infants may have the absence of characteristic signs seen in full‐term infants.

• Pyloric stenosis in premature infants may present out of the usual age range.

• A high degree of suspicion, and reliance on sonographic features may be required for diagnosis in this subgroup of patients, who may be misdiagnosed as having feed intolerance.

• Consider investigations early in the setting of an affected monozygotic twin.

1. Background

Pyloric stenosis occurs due to genetic and environmental factors, with an association of 0.25%–0.44% in monozygotic twins [1]. Bienfait et al. [1] reported the mean age of diagnosis at 38 days, though this may be underdiagnosed in premature infants as pyloric stenosis may present as feed intolerance in this group. Here, we present a case of monozygotic twins, who presented at 30 days old with feed intolerance (non‐bilious vomiting), normal blood gas, and ultrasound demonstrating hypertrophic pyloric stenosis—an atypical presentation of twins who developed pyloric stenosis simultaneously.

2. Case Presentation

Two monochorionic diamniotic twin males were born to a 35‐year‐old female, G6P0, at 32 + 4 weeks gestation, via in vitro fertilisation pregnancy. Non‐invasive prenatal testing was low risk, with 14% growth discordance, but no twin‐twin transfusion syndrome. Maternal history included short cervix requiring cerclage and progesterone, and iron deficiency. There was no known family history of pyloric stenosis. Delivery was via emergency caesarean section for an abnormal cardiotocogram.

Twin 1 was born with normal Apgar scores. Twin 2 required resuscitation and stimulation, as Apgar scores were 6 and 8 (1 and 5 min). They were admitted to the neonatal intensive care unit for respiratory distress syndrome and jaundice requiring phototherapy. Twin 1 required continuous positive airway pressure (CPAP) for 2 days. Twin 2 required CPAP for 6 days. Both were transferred to the special care nursery (SCN) at 7 days. Twin 1 was discharged to Hospital in The Home (HITH) at corrected age 35 + 5 weeks and discharged from HITH 2 days later after adequate weight gain. Twin 2 was discharged to HITH at corrected age 36 + 4 weeks. Both required nasogastric tube feeding during admission.

Twin 2 was re‐admitted to SCN 2 days later for non‐bilious vomiting post feeds, with a working diagnosis of gastro‐oesophageal reflux. He was commenced on omeprazole. However, due to persistent vomiting and slow growth, an abdominal ultrasound was requested. This showed features of pyloric stenosis, with 3.3 mm muscle wall thickness and 17 mm pyloric canal at corrected age 37 weeks (Figure 1). Only trickle flow of gastric contents was seen through the pylorus. Blood gas remained normal throughout admission (Table 1 showing admission results). He underwent laparoscopic pyloromyotomy on day 5 of admission; intra‐operative findings confirmed pyloric stenosis (Figure 2). He progressed well with feeding post‐operatively (formula) to be discharged home on day 7 of admission.

FIGURE 1.

Abdominal ultrasound images. (A) Twin 1 initial ultrasound. (B) Twin 1 repeat ultrasound. (C) Twin 2.

TABLE 1.

Capillary blood gas and electrolyte values for Twin 1 and 2.

	pH	pCO2	HCO3 (mmol/L)	Na (mmol/L)	K (mmol/L)	Cl (mmol/L)
Twin 1	7.38	45	24.1	141	4.7	107
Twin 2	7.40	46	25.1	144	4.4	1.8

FIGURE 2.

Intra‐operative images. (A) Twin 1. (B) Twin 2.

Due to Twin 2's readmission, parents were concerned Twin 1 may have similar diagnosis, though clinically he was having possets post feeding, without high clinical suspicion of pyloric stenosis. He was brought into emergency department at corrected age 36 + 6 weeks and observed overnight, with intermittent milky vomits. He had normal blood gas (Table 1). On examination, abdomen was soft and non‐distended with no palpable masses. Due to the twin's history, an abdominal ultrasound was obtained, suggesting evolving hypertrophic pyloric stenosis, with 3.3 mm wall thickness and 14 mm pyloric canal (Figure 1). Only trickle flow was seen through the pylorus. He had repeat ultrasound the following day, which was consistent with hypertrophic pyloric stenosis—3 mm muscle thickness and 15 mm pyloric canal, with no transit of fluid through the pylorus (Figure 1).

He underwent laparoscopic pyloromyotomy; intra‐operative findings confirmed pyloric stenosis (Figure 2). He progressed well with feeding post‐operatively (formula) to be discharged home on day 2 post‐operation, at the same time as Twin 2.

On follow up, both are thriving with no maternal concerns.

3. Discussion

Current evidence suggests that premature babies with pyloric stenosis may have absence of signs characteristic of full‐term infants, including absence of voracious appetite, projectile vomiting, and gastric peristalsis [2, 3]. While classic presentation of hypertrophic pyloric stenosis is between 3 and 6 weeks old, premature babies may present outside this range, even before corrected age of term. In our patients, this was observed, with absence of characteristic signs, and presentation before corrected age of term. Therefore, a high degree of suspicion and reliance on sonographic features may be required in this group.

A large retrospective study by Stark et al. [4] suggested that the incidence of pyloric stenosis in premature infants is 2.99 per 1000, while that in term infants is 2.25 per 1000 (relative risk 1.33, 95% confidence interval 1.16–1.54). In this study, 8.43% of patients were premature, while 91.57% were term. The median (IQR) chronological age at presentation was 40 days (30–56) in premature infants compared with 33 days (26–45) in term infants, while the median corrected gestational age was 42 weeks (40–42) for premature infants compared with 45 weeks (44–46) in term infants (p < 0.001).

Factors associated with pyloric stenosis include multiple pregnancies, male gender, and exposure to macrolides. Mendelian inheritance pattern has not been documented, even in monozygotic twins [5]. Other risk factors of pyloric stenosis include caesarean birth, maternal smoking during pregnancy, and first‐born. Our patients had risk factors of multiple pregnancy, male gender, and caesarean birth.

A study by Krogh et al. [6] demonstrated a higher incidence in monozygotic twins (200‐fold) compared with dizygotic twins (20‐fold), and newborns with an affected monozygotic twin were six times more likely to develop pyloric stenosis than newborns with an affected dizygotic twin. This supports a possible role of genetics in pyloric stenosis.

Abdominal ultrasound in pyloric stenosis is highly sensitive (91%) and specific (100%) [4]. It is considered not to be affected by weight, corrected gestational age or duration of symptoms in premature infants [7]. Certainly, there is increased dependency on ultrasound in current practice due to an increase in ultrasound utilisation and resolution, and a drive to prevent aspiration risk during test feed [8]. In our patients, reliance on sonographic features played a major role in diagnosis and management, as they did not present with characteristic signs. Further, we would recommend ultrasound of the pylorus as part of the work‐up for premature infants not progressing in weight gain associated with persistent feed intolerance.

The lack of biochemical changes in our patients likely reflects improved nourishment of infants nowadays, with literature suggesting that significant electrolyte derangements are no longer a consistent finding [9]. Our patients, while presenting without symptoms and signs characteristic of findings seen in term infants, highlight the need to have a degree of clinical suspicion and reliance on ultrasound in making the diagnosis of pyloric stenosis.

Ethics Statement

Ethics approval was obtained from Monash Health Research Office (ERM reference number 116819).

Consent

Informed consent was obtained from parents of the patients.

Conflicts of Interest

The authors declare no conflicts of interest.