Acute upper gastrointestinal bleed in the puerperium following weight reduction surgery

Abstract

Bariatric surgery is often a definitive treatment for obesity and is increasingly being performed on women of childbearing age. While bariatric surgery may reduce the risk of obesity in pregnancy, there are new complications which can develop following these procedures. Our case describes a 31-year-old women who presented in the puerperium with a life-threatening upper gastrointestinal bleed secondary to marginal ulceration following a Roux-en-Y procedure. This case report discusses a rare case of acute upper gastrointestinal bleed in the postnatal period and highlights the risk factors and complications which may present in the obstetrical patient following bariatric surgery. With the increasing use of weight loss surgery in obese women and the associated improvement in fertility following, we must remain aware of the risks and these women should be identified at booking so that their antenatal and postnatal care can be tailored accordingly.

Background

Bariatric surgery is a definitive treatment for obesity which is increasingly being used. Additionally, a large proportion of patients undergoing this procedure are women of childbearing age.^{1 2}

A late complication of bariatric surgery includes the development of a marginal ulcer which is defined as an ulcer at, or near to, the gastrojejunostomy (Roux-en-Y anastomosis). The incidence of this following bariatric surgery ranges from 0.6% to 16%^3–5 and can result in complications such as perforation and major haemorrhage secondary to upper gastrointestinal bleeding.⁶ There is little-to-no data reporting this complication in pregnancy and puerperium.

Acute massive upper gastrointestinal bleeding is a medical emergency requiring aggressive resuscitation and rapid intervention to control the source of bleeding and prevent serious morbidity and mortality. This case reports on a rare presentation of a life-threatening upper gastrointestinal bleed in the postnatal period and highlights the importance of the identification of risk factors antenatally to avoid acute complications.

Case presentation

A 31-year-old women, underwent an elective caesarean section at 37 weeks gestation due to a previous caesarean section, gestational diabetes and intrauterine growth restriction.

During her booking appointment, it was identified that she had previously undergone a Roux-en-Y gastric bypass for morbid obesity only 14 months previously. Due to her surgery, she took regular medications including omeprazole and forceval, and was required to take these lifelong. She conceived approximately 5 months following this procedure, and during this time period had lost approximately 30 kg, with a body mass index of 33.7 at booking. Her past medical history included postpartum psychosis and she was an ex-smoker, having quit in the year prior to conception. In this pregnancy, she had a low pregnancy associated plasma protein-A (PAPP-A), which required regular growth scans throughout the pregnancy. These growth scans showed evidence of intrauterine growth restriction, with an estimated growth to be less than the fifth centile. During the pregnancy she also developed gestational diabetes mellitus which was treated with insulin and metformin.

During her pregnancy, she had no issues with regards to her bariatric surgery; however, at 20 weeks gestation she was seen with acute upper abdominal pain which was attributed to gallstones. A plan was put in place for the patient to undergo a cholecystectomy following delivery.

At 37 weeks gestation, her caesarean section was performed without any complications, despite this, she remained in hospital post-delivery due to problems with urinary incontinence and neonatal jaundice. During her postoperative period, she stopped taking omeprazole as she felt this was making her abdominal pain worse. She was also prescribed regular ibuprofen which is routine obstetrical postoperative pain relief and dalteparin for prophylaxis against venous thromboembolism.

On her 5^th day post-delivery, she developed an acute gastrointestinal bleed, which presented as large volume haematemesis and a significant tachycardia.

Investigations

Baseline investigations were performed including full blood count, urea and electrolytes, liver function, group and save. A venous blood gas was performed which revealed a pH of 7.33 and a lactate of 2.0 mmol/L. The initial haemoglobin was 121 g/dL, this rapidly dropped to 86 g/dL over a 2-hour period. The clotting remained normal throughout with an activated partial thromboplastin time (APTT) ratio of 1.16, an international normalised ratio (INR) of 0.9 and fibrinogen was 3.55 g/L. The C-reactive protein (CRP) was 40 mg/dL and urea and electrolytes also remained normal with a urea of 3.3 mmol/L and an estimatedglomerular filtration rate (eGFR) maintained >90 mL/min/1.73 m².

Treatment

The patient was rapidly and aggressively resuscitated, the major obstetric haemorrhage protocol was activated and intravenous fluid commenced followed by O Negative blood. The patient was commenced on intravenous omeprazole, tranexamic acid, intravenous metoclopramide and intravenous erythromycin.

Due to the severity of the situation, she was transferred to the high dependency unit on delivery suite, she was subsequently reviewed by the obstetric anaesthetist, medical registrar and on-call gastroenterology consultant. Following a gastroenterology review, the patient was taken for an emergency gastroscopy under general anaesthetic. During this emergency procedure, an arterial bleed from a gastric ulcer at the site of the Roux-en-Y anastomosis was identified. This was treated with epinephrine and endoscopic haemoclips to secure haemostasis. The time from presentation to endoscopic treatment was 80 min. The total blood loss was estimated at approximately 2 L.

Outcome and follow-up

Following the procedure, the patient recovered well with regular review from the gastroenterology and obstetric team. She was fully debriefed following the events and her mood was closely monitored. A 3-day infusion of intravenous omeprazole was completed and, following this, buccal lansoprazole was commenced for 12 weeks. She remained on soft diet for 1 week following the events. She was subsequently discharged with a follow-up gastroscopy in 8 weeks to ensure ulcer healing.

Following this case, the bariatric service now advise all patients undergoing weight loss reduction surgery not to take non-steroidal anti-inflammatory drugs (NSAIDs) and if any patient reports pregnancy during the follow-up period, this advise is passed on to the obstetric team for the patients subsequent antenatal and postnatal care.

Discussion

The 2019 MBRRACE report identified that, of all direct maternal deaths, 34% of women were obese and a further 24% were overweight.⁷ Bariatric surgery is arguably the most effective treatment for the long-term management of obesity with 70% to 86%⁸ of women undergoing bariatric surgery being of childbearing age.^{1 9} With weight loss surgery improving fertility and reproductive outcomes in obese women, there is likely to be more women conceiving following such procedures and consequently more associated complications.¹⁰ While the Royal College of Obstetricians and Gynaecologists has issued guidance on the management of obesity in pregnancy,¹¹ the guideline does not address the management of pregnancy following bariatric surgery.^{9 11}

Bariatric surgery reduces the risk of obesity-related pregnancy complications such as gestational diabetes mellitus, pre-eclampsia and macrosomia,¹² however, it is also accompanied with its own risks such as small for gestational age in the fetus, anaemia, pregnancy loss and nutritional deficiencies.⁸ Recently better outcomes have been seen in those undergoing Roux-en-Y and laparoscopic adjustable banding although evidence is conflicting.⁸ A retrospective review Dao et al² compared pregnancy outcomes following gastric bypass surgery and found no significant difference in malnutrition, adverse mental outcomes or pregnancy complications. Patel et al⁸ specifically compared pregnancy outcomes following Roux-en-Y bypass and found similar complication rates with non-obese controls and lower complications than those who were obese and severely obese. A critical review performed by Guelinckx et al¹³ found few case and cohort studies, however these did show a reduction in obstetrical complications following gastric bypass. They also reported an increased risk of intrauterine growth restriction and nutritional deficiencies in both the mother and fetus.

The bariatric services at our hospital advise all women undergoing weight reduction surgery regarding their increased fertility following this procedure and advise the avoidance of pregnancy for at least 2 years following. Little is known about the influence of pregnancy timing on post bariatric surgical complications.¹² Current recommendations suggest waiting a period of 12 to 18 months prior to conceiving after bariatric surgery to ameliorate some of the risks.⁹ In particular, avoiding pregnancy during the rapid weight loss stage and during the period that most nutritional and electrolyte imbalances can occur,⁹ however, these recommendations are mostly based on theoretical risk.^{9 12} Very few studies have evaluated the influence of postoperative interval on pregnancy outcomes or postoperative complication rates, however, in those studies that have been performed, no difference was found in the rates of adverse outcomes including bariatric complications in those who conceived during versus after the first postoperative year.¹²

The Roux-en-Y gastric bypass is among one of the most commonly used bariatric procedures.⁹ Longer-term complications of Roux-en-Y gastric bypass include marginal ulceration at the gastrojejunostomy junction with an estimated incidence of 0.6% to 16%.^{3 4 6} These may present with perforation or massive bleeding in a patient with previously minimal symptoms, as seen in our case.⁶ Coblijn et al^{3 6} and Azagury et al¹⁴ identified risk factors that were significantly associated with the development of marginal ulcers post procedure including the use of NSAIDs, along with smoking and diabetes. These risk factors were all seen in our case, and it is felt that the use of NSAIDs as pain relief in the postpartum period were the major contributing factor to the development of the anastomatic ulcer and subsequent ensuing major haemorrhage.

In the discussed case, the time from first presentation to endoscopic treatment was approximately 80 min. Due to the rarity of this scenario within the context of the maternity setting, there was a small delay in obtaining senior medical and gastroenterological review. Once the patient had been reviewed by the correct speciality, she was rapidly transferred to theatre, however this delay likely contributed to the total blood loss.

Treatment revolves around reducing the risk factors, commencing a proton-pump inhibitor and endoscopy, if indicated.⁴ The optimal endoscopic treatment advocated to treat a bleeding ulcer is by dual therapy with epinephrine infiltration and mechanical haemostats with endoclips, this is the treatment received in our case.^{15 16}

Patient’s perspective.

When the blood came, there was no pain; I was just unable to eat prior to the incident which I now know was due to the burst ulcer forming vessels to build blood clots in the digestive system as a result of receiving ibuprofen post C-section and not taking omeprazole. I’m so grateful to all the teams, nurses, midwives and doctors that saved my life. Thank you.

Learning points.

This report highlights a rare case of major upper gastrointestinal bleed in the puerperium following weight reduction surgery with a good outcome.

• With the increasing use of weight loss surgery in obese women and the associated improvement in fertility following, we must remain aware of the risks of this and the possibly life-threatening complications which may present in the antenatal and postpartum period.

• These women should be identified at booking so that their antenatal and postnatal care can be tailored accordingly, including the avoidance of ibuprofen and other non-steroidal anti-inflammatory drugs.

• Rapid diagnosis and resuscitation with a multidisciplinary team approach is key to managing the patient and ensuring a good outcome.