Surgery Now or Surgery Later? What New Research Says About Timing the Operation for Hirschsprung Disease

Three studies from Japan, Qatar and Vietnam look at whether newborns do better having their bowel operation in the first month of life or a few months later

Babies born with Hirschsprung disease need an operation to remove the section of bowel that cannot move stool along. Surgeons can do it in the first month of life or wait a few months. Three studies published in 2025 and 2026 found that both timings are safe, but that they trade different things: operating early means less total time in hospital during the first year, while waiting means an easier few months right after surgery.

What Hirschsprung disease is, and why timing became a question

In every healthy bowel there are nerve cells, called ganglion cells, embedded in the bowel wall. They are what makes the bowel squeeze in a coordinated wave to move stool along. In Hirschsprung disease, those nerve cells never finish forming in the last part of the bowel during pregnancy. That final segment stays permanently tight, stool cannot get past it, and the bowel above swells up. It affects roughly one baby in every 5,000, and it is usually picked up in the first days of life when a newborn does not pass meconium — the first, dark stool — and the abdomen becomes distended.

The treatment is an operation called a pull-through: the surgeon removes the segment without nerve cells and joins healthy, working bowel down to the anus. What has been genuinely unsettled is when to do it.

For most of the twentieth century the answer was to wait. The early operations, developed in the 1950s and reported with long follow-up in 1975, were risky enough that surgeons recommended waiting until at least four months of age, and in the meantime creating a stoma — an opening on the abdomen where stool is collected into a bag — to protect the baby [1]. That began to change in 1980, when a team reported doing the pull-through in newborns without a stoma first [2]. Over the following decades, operating techniques became far gentler: much of the work can now be done through the anus itself, or through a keyhole camera and one or two tiny abdominal incisions. As the operation got safer, surgeons moved earlier. In the United States, the average age at the operation dropped from six months in 1997 to one month by 2006 [3].

Here is the problem: practice moved before the evidence did. European guidelines for the commonest form of the condition deliberately did not name a best age [4], because the studies that existed disagreed and none of them could separate the effect of when the operation was done from the effect of which babies were chosen for it — sicker babies tend to be operated on sooner, which makes early surgery look worse than it is.

Meanwhile, waiting is not free either. In one group of children awaiting the operation, 17% developed a serious complication, including bowel perforation — a tear in the bowel wall [5]. A separate report found perforations in 6% of newborns on the waiting list [6]. And one study found that operating early cut the rate of a dangerous bowel infection called Hirschsprung-associated enterocolitis from 43.4% before surgery to 20.2% afterwards [7]. Families waiting at home also have real work to do: most babies need daily rectal irrigations, where a small tube and syringe are used to wash stool out of the bowel, until the operation happens.

The three new studies

The first study, from the University of Tokyo, used a national hospital database that covers 88% of Japanese hospitals with newborn intensive care units, spanning July 2010 to March 2021 [8]. The researchers found 365 babies who were diagnosed in the first month of life and had their pull-through within four months without a stoma first. Sixty-five had the operation within the first 30 days; 300 had it between 31 and 120 days. Crucially, they used a statistical method that mathematically balances the two groups on things like birth weight, prematurity, other birth defects and which type of hospital treated them — an attempt to imitate what a randomised trial would have achieved.

The second study, led from Sidra Medicine in Qatar with collaborators in Canada, was a systematic review and meta-analysis — a study of studies [9]. The team searched six medical databases from the earliest available records through June 2024, screened just over 5,000 papers, and combined the 20 that compared babies operated on before one month with children operated on later. Together those studies covered 3,197 patients: 1,371 newborns and 1,826 older children. Importantly, none of the 20 was a randomised trial.

The third study came from the Vietnam National Hospital of Pediatrics in Hanoi [10]. It followed 55 newborns operated on within their first 28 days, all by the same surgical team, and reassessed them four years or more later — long enough to ask the question families most care about: how is the child's bowel working now?

What they found

The Japanese study found that neonatal surgery was safe. The rate of complications during the hospital stay was no different between babies operated on early and those operated on later. The early group did stay longer after their operation — about 6.5 days longer — and that admission cost more. But when the researchers added up all the hospital days across the whole first year of life, the early group spent 14.6 fewer days in hospital than the later group, and the total first-year cost was no different. Their conclusion was direct: neonatal surgery is safe, and delaying may not benefit the baby [8].

The meta-analysis reached a different headline. It found that operations on newborns were quicker — about 26 minutes shorter on average — but that newborns had a harder first three months afterwards. They were roughly five times more likely to have sore, broken perianal skin at three months, about five times more likely to have an episode of enterocolitis in that window, and about three times more likely to develop a narrowing at the join in the bowel. Rates of wound infection, readmission within 30 days, and death were the same in both groups [9]. The authors' own honest caveat is important: every study they pooled was observational rather than randomised, the definitions of complications varied between them, and one of their headline numbers only became statistically significant after a single small study was removed.

They also demonstrated exactly the problem described earlier. At first it looked as though newborns had more enterocolitis before surgery. When the researchers removed the studies in which enterocolitis was itself the reason for operating early, the difference disappeared. In other words, part of what looked like a harm of early surgery was really a sign that sicker babies were being operated on sooner.

The Vietnamese study supplies the longer view. At an average of four and a half years after surgery, bowel function was good, and it was equally good whichever of the three minimally invasive techniques had been used. There were no deaths and no life-threatening complications. About one in four children had a minor complication and about one in eight had a more significant one needing a further procedure; enterocolitis occurred in 18% [10]. Another study found that newborns simply take longer to settle into a normal stooling pattern — about 11 weeks, compared with about 7 weeks for older children [11] — which fits the picture of a bumpier start that then evens out.

What this means for families

Put together, these studies say something reassuring: there is no evidence that either timing is dangerous. Deaths and major complications were the same either way. What differs is the shape of the first few months.

Choosing surgery in the newborn period means the operation is behind you sooner, no more daily irrigations at home — a burden that researchers are increasingly examining in its own right [12] — and, on the Japanese figures, roughly two fewer weeks in hospital across the whole first year. The cost is a longer stay right after the operation and a genuinely harder stretch of about three months: more frequent stools, sore skin around the bottom, a higher chance of an enterocolitis episode, and a possible narrowing at the surgical join that may need stretching.

Choosing to wait means a gentler recovery when the operation does happen, but weeks or months of irrigations at home first, more trips to hospital later in the year, and a small but real risk of a serious problem while waiting.

If your baby is having surgery in the newborn period, it is worth knowing in advance that frequent stools and sore skin in the first months are expected and temporary, not a sign the operation failed. Barrier creams and diligent skin care from day one make a real difference, and any fever, swollen abdomen or foul-smelling diarrhoea should prompt an urgent call, because enterocolitis needs treating quickly.

It also matters where the surgery happens. In the Japanese data, newborn operations were concentrated in university and high-volume hospitals [8], and the Vietnamese results come from one experienced team operating on a carefully selected group [10]. It is entirely reasonable to ask your surgeon how often their unit performs this operation on newborns.

What researchers are working on next

The honest summary is that no randomised trial has ever compared the two timings, and all three research teams say one is needed. What would help most is a large study that treats age at surgery as a sliding scale rather than a before-or-after-one-month split, that agrees in advance on how complications are defined, and that follows children into school age — because the outcome that matters most to a family is not the length of the first hospital stay but whether their child has normal bowel control at six, ten and sixteen. The longer-term evidence that does exist is encouraging: bowel function after this operation improves steadily as children grow and, by adulthood, comes close to that of people who never had the condition [13], and a large registry study found similar long-term function whether the operation was done early or later [14]. Until better evidence arrives, both choices remain reasonable — which means this is a decision to make together with your surgical team, in the light of your baby's own situation, rather than one with a single right answer.

References

  1. Swenson O, Sherman JO, Fisher JH, Cohen E. The treatment and postoperative complications of congenital megacolon: a 25 year followup. Ann Surg. 1975;182(3):266–273. doi:10.1097/00000658-197509000-00008 ↩
  2. So HB, Schwartz DL, Becker JM, Daum F, Schneider KM. Endorectal "pull-through" without preliminary colostomy in neonates with Hirschsprung disease. J Pediatr Surg. 1980;15(4):470–471. doi:10.1016/s0022-3468(80)80755-x80755-x) ↩
  3. Huang EY, Tolley EA, Blakely ML, Langham MR. Changes in hospital utilization and management of Hirschsprung disease: analysis using the kids' inpatient database. Ann Surg. 2013;257(3):371–375. doi:10.1097/SLA.0b013e31827ee976 ↩
  4. Kyrklund K, Sloots CEJ, de Blaauw I, et al. ERNICA guidelines for the management of rectosigmoid Hirschsprung's disease. Orphanet J Rare Dis. 2020;15(1):164. doi:10.1186/s13023-020-01362-3 ↩
  5. Beltman L, Labib H, Oosterlaan J, van Heurn E, Derikx J. Risk factors for complications in patients with Hirschsprung disease while awaiting surgery: beware of bowel perforation. J Pediatr Surg. 2022;57(11):561–568. doi:10.1016/j.jpedsurg.2022.02.022 ↩
  6. Liu Z, Zhang Y, Sun D, et al. Bowel perforation in neonates with Hirschsprung disease: a case series and literature review. Pediatr Surg Int. 2024;41(1):15. doi:10.1007/s00383-024-05907-6 ↩
  7. Xia R, Ma T, Zhao F, et al. The efficacy analysis of laparoscopic-assisted Soave procedure for Hirschsprung's disease in neonates and infants within 3 months of age. Sci Rep. 2024;14(1):31924. doi:10.1038/s41598-024-83422-3 ↩
  8. Kutsukake M, Aso S, Konishi T, et al. Perioperative outcomes of neonatal versus delayed surgery for Hirschsprung disease: a nationwide retrospective cohort study in Japan. Pediatr Surg Int. 2025;41(1):211. doi:10.1007/s00383-025-06126-3 ↩
  9. Alshahwani N, Alsaied A, Tewfik S, et al. Perioperative outcomes in primary neonatal pullthrough versus pullthrough in older children with Hirschsprung disease: a systematic review and meta-analysis. Sci Rep. 2026;16(1):7004. doi:10.1038/s41598-026-35690-4 ↩
  10. Tran QA, Pham HD, Ly DB, et al. Safety and feasibility of one-stage neonatal approach for short-segment Hirschsprung's disease. PLoS One. 2026;21(1):e0341212. doi:10.1371/journal.pone.0341212 ↩
  11. Kim HY, Oh JT. Stabilization period after 1-stage transanal endorectal pull-through operation for Hirschsprung disease. J Pediatr Surg. 2009;44(9):1799–1804. doi:10.1016/j.jpedsurg.2008.10.070 ↩
  12. Bokova E, Prasade N, Lewis WE, et al. Evaluation of post-neonatal intensive care unit home irrigations prior to pull-through: implications for Hirschsprung disease management. J Pediatr Surg. 2024;59(7):1245–1249. doi:10.1016/j.jpedsurg.2024.03.012 ↩
  13. Neuvonen MI, Kyrklund K, Rintala RJ, Pakarinen MP. Bowel function and quality of life after transanal endorectal pull-through for Hirschsprung disease: controlled outcomes up to adulthood. Ann Surg. 2017;265(3):622–629. doi:10.1097/SLA.0000000000001695 ↩
  14. Roorda D, Verkuijl SJ, Derikx JPM, et al. Did age at surgery influence outcome in patients with Hirschsprung disease? A nationwide cohort study in the Netherlands. J Pediatr Gastroenterol Nutr. 2022;75(4):431–437. doi:10.1097/MPG.0000000000003550 ↩