Bleeding and Dehydration: Two Registers of Early Risk After Bariatric Surgery

Ask what goes wrong early after a sleeve gastrectomy or a gastric bypass, and the honest answer is that you are asking about at least two different problems. Two national registry analyses published weeks apart describe them from opposite ends: the most serious bleeding complication appears largely indifferent to how the operation was performed, while the leading cause of readmission has been cut roughly in half in under a decade — and still holds first place.

Both are observational studies of administrative and audit databases. They are good at describing who ends up in trouble and how often. Neither can tell anyone which technique to choose.

The bleed that technique does not explain

The Dutch Audit for Treatment of Obesity followed 83,580 patients who had a primary metabolic bariatric operation between 2015 and 2024 — 18,701 sleeve gastrectomies and 64,879 bypass procedures, a surgical mix far more bypass-weighted than American practice. Post-operative haemorrhage of Clavien-Dindo grade 2 or higher, meaning bleeding that required something more than observation, occurred within 30 days in 1.2% of patients: 1,021 people. The authors note it is the most common severe short-term complication in this field, and that most of these bleeds needed a surgical reintervention.

Four variables emerged as independently associated with that outcome in a multilevel logistic model. A history of stroke or transient ischaemic attack carried an adjusted odds ratio of 1.75; liver disease, 1.84; hypertension, 1.34. The fourth points the other way. Each additional BMI point was associated with slightly lower odds of bleeding, an adjusted ratio of 0.974.

Conspicuously absent from that list is anything the surgeon controls. Stapler brand, stapling technique, tranexamic acid protocols and thromboprophylaxis regimens were all collected through nationwide questionnaires specifically to be tested as candidate factors. The paper’s own summary states there was no association between surgical techniques and post-operative haemorrhage, and patient-level variables are the only ones it reports as independently associated.

What an inverse BMI signal is not

It is worth being precise about that 0.974. It is an association measured across a population, expressed per BMI unit, in a retrospective audit in which the clinical detail was gathered after the fact from questionnaires — a design the authors themselves flag as a limitation. It is not a threshold, not a mechanism, and not something a person could act on. The more defensible reading is that bleeding risk here behaves less like a technical artefact and more like a marker of underlying vascular and hepatic health, which is a much harder thing to fix on the day of surgery than to identify beforehand.

The readmission that has quietly halved

The second analysis, published in Surgery, queried the Metabolic and Bariatric Surgery Accreditation and Quality Improvement Program database — the same national registry we examined earlier this week for procedure volumes — and pulled 1,456,480 adults who had a primary minimally invasive sleeve or Roux-en-Y bypass between 2015 and 2023. Mean age was 44, mean BMI 45.2, and 73.2% of the operations were sleeves, almost exactly inverting the Dutch proportions.

Thirty-day readmission ran at 3.3% overall. Of those readmissions, 28% were coded for nausea, vomiting, fluid, electrolyte or nutritional depletion, renal insufficiency or renal failure — the cluster the authors group as dehydration and acute kidney injury. That is close to one readmission of this kind for every hundred patients operated on.

Looked at year by year, the trend runs the right way. Readmission for dehydration or kidney injury fell from 1.2% to 0.5% after sleeve gastrectomy and from 1.8% to 0.9% after gastric bypass, both changes with a p-value below .0001. The improvement is real and large.

Who still comes back

Halving a rate does not dethrone it: dehydration and kidney injury remain the single largest readmission category in the database. And the factors associated with persistently higher odds are worth stating plainly, because they are not distributed evenly. Sleeve gastrectomy carried higher odds than bypass — the opposite of what the bleeding analysis might lead you to expect about the “simpler” operation. Female sex carried higher odds. So did a diagnosis of insulin-dependent diabetes, hypertension or renal insufficiency. So did being recorded as Non-Hispanic Black.

That last association is a registry signal, not an explanation. The authors frame it as identifying high-risk populations and pointing to a need for targeted preventive strategies, and the study design cannot separate biology, comorbidity burden, insurance, distance from care or follow-up intensity. Studies such as the LABS study, funded by NIDDK, were built prospectively for exactly that kind of question — see the LABS study archive — and the honest position is that the readmission gap is documented better than it is understood.

Where the leverage sits

Put the two papers side by side and the pattern is uncomfortable for anyone hoping for a technical fix. The Dutch cohort shows the most feared early bleed is tied to the patient’s vascular and liver history rather than to staplers or drugs. The American cohort shows the most common reason for coming back has improved dramatically without any corresponding surgical breakthrough, and now concentrates in identifiable groups. Neither finding is a discovery of cause; both are descriptions of where risk lives.

Which leaves the pre-operative assessment and the days after discharge doing most of the work — and the baseline BMI, the number that opens the whole conversation, still plotted through something as ordinary as our BMI calculator. The archive at niddklabs.org is independent and unaffiliated with NIH or NIDDK; it reports what registries and journals publish, and nothing here is a recommendation for or against surgery for any individual.