Laparoscopic Pancreatectomy for Persistent Hyperinsulinemic Hypoglycemia of Infancy – Case Series

ABSTRACT

Context:

Hypoglycemia in neonates can be attributed to Persistent Hyperinsulinemic Hypoglycemia of Infancy (PHHI). Even though initial management involves the use of medical agents, requirement of surgery is seen in a large number of cases, particularly in those who are nonresponsive. Open pancreatectomy has been the gold standard in the management of PHHI, but there has been a recent inclination toward the use of minimally invasive procedures, namely laparoscopic pancreatectomy to achieve similar results with lesser morbidity.

Aims:

The aim of this study is to describe the presentation, key technical points, and outcomes of laparoscopic pancreatectomy in the management of PHHI.

Materials and Methods:

A retrospective study which was conducted between January 2019 and September 2024 which included 7 children who were diagnosed to have PHHI.

Results:

All seven patients who were operated were infants with an average age of 4 months (range 1–5 months), with the youngest patient being one and half months of age. All of them underwent a laparoscopic pancreatectomy with the average duration of the procedure being 60 min and with very minimal blood loss. A distal pancreatectomy was performed in one infant while a near total pancreatectomy was performed in the other six infants. Euglycemia was achieved in six cases and they are on regular follow-up, with the longest duration of follow-up being 4 years and 3 months. The average duration of follow-up has been 3 years and 6 months, ranging from 1 month to 4 years and 3 months. One child had postoperative requirement of octreotide for 2 weeks after which it was tapered and stopped. One child had persistent hypoglycemic episodes and required a completion pancreatectomy 6 months after the initial surgery. No intraoperative or postoperative complications were noted.

Conclusion:

Laparoscopic pancreatectomy is an excellent tool in the armamentarium of surgeons looking for an effective surgical modality to treat PHHI with minimal morbidity.

INTRODUCTION

Persistent Hyperinsulinemic Hypoglycemia of infancy (PHHI) is a rare derangement of glucose metabolism, which carries an estimated incidence of 1–1.4 in 50,000 live births. Early diagnosis is essential as it causes severe hypoglycaemia and may lead to permanent brain damage to neonates and children.[1] Conventionally, open pancreatic resection was done for PHHI in infants who are refractory to medical management. Advancements in minimally invasive surgery have enabled surgeons to perform laparoscopic pancreatectomy safely in infants.[2] While the need for pancreatectomy for PHHI is well documented, there is a definite dearth of literature regarding the technical aspects of performing a laparoscopic pancreatectomy and its efficacy in the management of PHHI.

Aim

The aim of this study is to describe the presentation, key technical points, and outcomes of laparoscopic pancreatectomy in the management of PHHI.

MATERIALS AND METHODS

A retrospective study was done between January 2019 to September 2024 which included 7 patients with PHHI who underwent laparoscopic pancreatectomy (laparoscopic distal pancreatectomy or a near total pancreatectomy) in our institute. Data collected included demographic details, fasting blood sugar levels, serum insulin levels, serum C peptide and Betahydroxybutyrate levels, Gallium 68 14,710 – Tetraazacyclododecane-14,710-tetraacetic acid-1-Nal3-Octreotide (DOTANOC) positron emission tomography (PET) scan reports, genetic study report, procedure details, and complications.

Diagnostic criteria of PHHI[1] was established. The criteria included:

• Fasting hypoglycemia <50 mg/dl,

• Fasting serum insulin level >2.0 µU/ml

• Low plasma hydroxybutyrate <1.8 mmol/L

• Free fatty acids (<1.7 mmol/L).

Only those children who fulfilled the biochemical, imaging and genetic criteria for diagnosis of PHHI were labeled as having PHHI.

Computed tomography brain was done to rule out neurological problems

The initial medical management of PHHI involved maintaining euglycemia by frequent oral feeding or orogastric feeds, continuous infusion of intravenous glucose, and administration of diazoxide, glucagon, and octreotide. Infants who responded to oral diazoxide therapy were worked up for genetic analysis (K-ATP, GDH, and GCK) and cases where failure to respond to medical therapy was noted, localization of PHHI was done with Gallium68 DOTANOC PET/computed tomography (CT). While a genetic analysis to detect specific mutations was not done for the initial 3 children due to logistic reasons, the last four children who underwent laparoscopic near total pancreatectomy had a genetic analysis done and a mutation involving ABCC8 gene was documented. Failure to respond to medical management was consider as an indication for surgical intervention in the form of laparoscopic pancreatectomy. Postoperatively, these infants were kept nil per oral and started on a Glucose Infusion Rate (GIR) of 4–6 mg/kg/min with hourly blood glucose monitoring. Oral feeds were resumed after 48 h and GIR was tapered and stopped. Drain tube was removed on the postoperative day 3 and the parents were educated regarding regular blood glucose monitoring. These infants were discharged home on the postoperative day 7. All infants except the last child were followed up for 1 year with initial weekly outpatient visits for 2 months and followed by once a month in the pediatric gastroenterology and pediatric endocrinology OPD, to look for exocrine and endocrine pancreatic insufficiency. In the pediatric endocrinology OPD, HbA1c was done every 6 months and the parents where trained to monitor the child’s blood glucose levels using a glucometer every month. In the pediatric gastroenterology OPD, clinical assessment was done including history of steatorrhea and growth monitoring and Fecal Elastase -1 level was measured every 6 months.

Surgical technique

Near total pancreatectomy

We used a three-port technique, one 5-mm camera port (supraumbilical) along with, two 5 mm working ports in the right and left lumbar regions [Figure 1a]. Two stay stitches were taken on the stomach and hitched to the anterior abdominal wall for retraction. The 2^nd stay stitch was taken close to the pylorus, thus leading to a stretching of the duodenum and better visualization of the pancreatic head and neck. Gastrocolic omentum was then divided to enter the lesser sac. The tail of the pancreas was identified first and the splenic vessels were dissected at this level. The splenic vessels are more slack and distinct at the level of the tail of the pancreas close to the hilum. It is easier to dissect the splenic vessels here and then proceed medially. The tail of the pancreas was then dissected and divided first. The pancreas was then dissected proximally, separating it from the splenic vessels and meticulously dividing its branches to the pancreas until the neck of the pancreas. Further dissection was limited by the superior mesenteric vessels. Harmonic scalpel was used to transect the pancreas (duct and parenchyma) at this level [Figure 1b]. Branches from the inferior pancreatico-duodenal vessels were identified and divided. The uncinate process and portion of the head of the pancreas were divided and removed as piece meal. The division of the head of pancreas was done 1 cm to the right of the superior mesenteric vein (SMV) to avoid injuring the common bile duct [Figure 1c and d]. Drain was placed in the lesser sac. Specimen was retrieved by widening the 5-mm camera port.

Figure 1.

(a) Port position, (b) division of the tail of pancreas, (c) division of head of the pancreas, (d) post near-total pancreatectomy showing the bared out superior mesenteric vein and splenic vessels

RESULTS

We had seven infants who were admitted in our pediatric intensive care unit with seizures and recurrent hypoglycemic events, fulfilling the above-mentioned diagnostic criteria [Table 1].

Table 1.

Investigation profile

Investigations	Case1	Case 2	Case 3	Case 4	Case 5	Case 6	Case 7
Fasting blood glucose (mg/dl)	<50	<50	<50	<50	<50	<50	<50
Serum insulin levels (µ/mL)	4.2	3.6	5.4	5.0	4.1	4.9	5.2
C peptide levels (ng/mL)	3.0	4.4	3.2	4.1	3.2	4.0	4.8
Beta hydroxybutyrate levels (mmol/L)	1.8	2.0	1.7	1.6	1.9	1.5	1.5

A genetic analysis to detect specific mutations was not done for the initial 3 children due to logistic reasons, the last 4 children who underwent laparoscopic near total pancreatectomy had a genetic analysis done and a mutation involving ABCC8 gene was documented.

A PET-CT (68 Ga DOTANOC scan) [Figure 2] was done for localization, which revealed a focal lesion in the tail of pancreas in 1 infant while the other 6 had diffuse lesion involving the neck, body, and tail of the pancreas.

Figure 2.

Diffuse uptake in the body and tail of pancreas

Parents were counseled regarding the need for surgery and the possibility of developing pancreatic exocrine insufficiency and diabetes mellitus at a later need and appropriate consent was obtained.

These seven infants, with a diagnosis of focal and diffuse PHHI, underwent laparoscopic distal pancreatectomy (Case 1) and laparoscopic near total pancreatectomy (Case 2, 3, 4, 5, 6, 7), respectively. The mean age at procedure was 5.16 ± 2.8 months with the mean weight at the time of procedure being 4.72 ± 1.6 kg [Table 2].

Table 2.

Operative details

Patient	Age	Weight (kg)	Procedure done	Resection extent (%)	Length of procedure (min)
1	45 days	2.9	Laparoscopic distal pancreatectomy	70	65
2	5 months	5.8	Laparoscopic near total pancreatectomy	90	75
3	1 month	4.1	Laparoscopic near total pancreatectomy	90	70
4	28 days	3.8	Laparoscopic near total pancreatectomy	90	60
5	40 days	4.3	Laparoscopic near total pancreatectomy	90	75s
6	7 months	7.4	Laparoscopic near total pancreatectomy	90	70
7	2 months	5.6	Laparoscopic near total pancreatectomy	90	75

Extent of pancreatectomy was 70% - distal pancreatectomy (case 1) and 90% - near total pancreatectomy (Case 2–7). Our mean operative time was 65.8 ± 7.3 min with negligible blood loss. None of them had any postoperative complications such as bile duct injury, pancreatic duct leak, and wound infection. The first child who was operated is now 5 years and 3 months of age and has not developed any features of pancreatic exocrine insufficiency or diabetes mellitus. One infant required postoperative octreotide injections for 2 weeks after which it was tapered and stopped. One child had persistent hypoglycemic episodes in the postoperative period, which could not be controlled with medications and required a completion pancreatectomy 8 months later. Six of these infants were euglycemic postoperatively in spite of stopping octreotide and oral diazoxide for >18 h. Histopathological study showed abnormal islet cell with nucleomegaly[11] consistent with PHHI [Figure 3].

Figure 3.

Abnormal islet cells with nucleomegaly

The mean duration of the follow-up has been 3 years and 6 months, ranging from 1 month to 4 years and 3 months.

None of them had any hypoglycemic events and thrived well with normal weight gain on follow-up. None of the operated infants have had episodes of hyperglycemia or features suggestive of exocrine insufficiency.

DISCUSSION

PHHI is a very rare disease of glucose metabolism which is thought to be due to an abnormal regulation of insulin production by pancreatic beta islet cell. The management of PHHI has advanced significantly over the past few decades with the development of effective pharmacological options and better surgical options. It is to be noted that medical management is not documented to be very successful with a success rate of 24% to 86%.[2]

The role of Gallium – 68 DOTANOC scan to localize and stratify PHHI as being focal or diffuse is very important. 18F L-DOPA PET/CT scan is more accurate than Gallium68 DOTANOC PET/CT scan,[3] but it is not available in Southern India at present.

Genetic analysis to document the presence of mutations involving specific genes, namely ABCC8 and KCNJ11 has been found to have a profound impact in planning the line of management in PHHI. ABCC8 gene codes for the Sulfonyl urea receptor 1 regulatory. subunit of the K–ATPase channel while KCNJ11gene encodes for the Kir6.2 (Potassium inward rectifying pore) subunit of the K– ATPase channel. Identification of diffuse and focal types of PHHI can be done using genetic studies. Single heterozygous recessive mutation in either ABCC8 or KCNJ11 gene has a 94% positive predictive value for focal disease while two recessive mutations may point toward a diffuse disease. Genetic analysis can also help is predicting response to medical management. Most patients with PHHI having ABCC8/KCNJ11 are diazoxide unresponsive and require near total pancreatectomy.[4,5]

Laparoscopic pancreatectomy poses a serious challenge to even the most experienced of laparoscopic surgeons. Its retroperitoneal location, preserving the splenic vessels and the spleen, handling the pancreatic duct and uncinate process and deciding the extent of dissection are some of the challenges encountered.

Retraction of the stomach

Entry into the lesser sac and visualization of the pancreas is best done by retracting the stomach using either a single trans cutaneous suture[6] or 2 transabdominal.[7] In our case series, we resorted to taking two stay stitches, one close to the pylorus and one on the body of the stomach [Figure 3].

Preservation of the splenic vessels and spleen

It is very important that the splenic vessels and the spleen be preserved for its obvious role in maintaining immunity against various organisms, especially in distal pancreatectomy. Although there are various techniques described to preserve the spleen, one of the most important techniques are the Warshaw technique and the Kimura technique. Kimura technique involves dividing all the branches from the splenic vessels entering the pancreas while preserving the splenic vessels themselves while Warshaw technique involves resection of the splenic vessels and allowing the spleen to be vascularized by the short gastric and Gastro epiploic vessels. Kimura’s technique has been reported to provide a better blood supply to the spleen but is more technically demanding when compared to Warshaw technique.[8] We adopted the Kimura technique in all our patients wherein we identified the branches to the pancreas, which we cauterized and divided taking care to preserve the main splenic vessels.

Duct and parenchyma

Division of the parenchyma and duct can be done in laparoscopic pancreatectomy by numerous techniques like using a linear stapler, ultrasonic shears, electrocautery, and suture ligation. The most dreaded complication associated with the division of the parenchyma and the duct is the formation of a pancreatic fistula. DISPACT Trial did not show any significant difference in the occurrence of pancreatic fistula among the hand sewn closure group when compared to the stapler group.[9]

Suzuki et al. have reported the reduced incidence of fistula formation if parenchyma and duct division was done using ultrasonic dissectors.[10,11] Suture line reinforcement can be done with fibrin glue, polypropylene mesh, or seromuscular patch[12,13,14]

Comparison between the various methods of closure of the pancreatic parenchyma and duct has shown little to no superiority of one over the other. Cecka F et al. have speculated in their meta-analysis that the perfection of a technique in an institution or by an individual surgeon is just as important as the actual technique applied.[15]

All these studies have been done on adults, and hence, there is a definite need for a randomized control trial to assess the feasibility and efficacy of these techniques in the pediatric population.

Division of the duct and parenchyma was done using an ultrasonic shear in all our patients and no attempt was made to ligate the duct separately.

Uncinate process

Near total pancreatectomy involves the removal of the uncinate process. The main hindrance to remove the uncinate process is the superior mesenteric vessels. In our case, we resorted to removal of the uncinate process as piece meal. Another method described in the literature to remove the uncinate process is the hanging maneuver of the mesenteric root as described by Pessaux P et al. in adults. In the hanging maneuver, a tape is passed around the SMA from its origin from the aorta thus improving visibility permitting adequate dissection.[16] There are no references available in the literature regarding any specific method of removal of the uncinate process in children.

Extent of resection

What is the extent of surgical resection of the pancreas is a question which has no definitive answer. While radical surgical resection (>95%) has been advocated by some to bring about adequate control, others prefer a more conservative approach to minimise the risk of postoperative diabetes mellitus.

In this tug of war between the radicals and conservatives, there are arguments which support both. The former feel that radical resection is an effective modality in controlling hypoglycemic episodes with possibilities of postoperative hypoglycemia being low.

The later have, however challenged these assumptions by showing that 1/3^rd of these patients who underwent radical resection had persistent hypoglycemia and 2/3^rds of them developed diabetes mellitus at a later date.[17]

Hence, the localization of the lesion is now being considered as a prerequisite to ensure curative resection. Some authors recommended partial pancreatectomy for focal lesions and near total pancreatectomy for diffuse variant based on an intraoperative pathological stratification. However, it is to be remembered that even though the use of intraoperative pathological examination of frozen sections may seem logical, it is seldom possible to differentiate with certainty the variants of PHHI.[18] This variability in reporting can subsequently result in variable resection being done for PHHI.

The extent of surgical resection is also dependant on the experience of the surgeon in handling this rare surgical condition.[19]

Extensive pancreatectomy (>95%) may not always be warranted as the first procedure in most of the patients due to various reasons. First one is the availability of effective medications which are capable of controlling hypoglycemic episodes. Second reason being the persistence of hypoglycemic episodes irrespective of the extent of resection. Third and an important reason was the documented incidence of early diabetes mellitus.[19]

There is a definite dearth of information regarding using laparoscopy to perform a pancreatectomy, more so for near total pancreatectomy. Pierro A et al. reported one of the largest series of laparoscopic near total pancreatectomy (22 cases) for medically unresponsive PHHI. However, it is to be noted that the procedure was completed laparoscopically only in 15 cases with 7 conversions to open surgery with the indication being bleeding. They reported bile duct injuries in 3 of their cases and 2 children diabetes mellitus and pancreatic insufficiency.[20]

Al - Shanafey S et al. reported a series of 12 cases of laparoscopic near total pancreatectomy with 2 children requiring a conversion to open surgery. The extent of pancreatectomy was lesser in the above-mentioned series (85%–95%) which contributed to higher incidence of postoperative hypoglycaemic episodes[21] A comparative study between laparoscopic pancreatectomy versus open for PHHI done by Al-Shanafey S et al. noted a more extensive resection being done by open technique than by laparoscopy (93% vs. 90%).[22] Onset of orals was found to be significantly earlier in the laparoscopic group with similar outcome.[21]

Laparoscopic pancreatectomy is a technical challenge, especially when dissecting the head of pancreas and the uncinate process due to possible injury to vital vascular structure, namely the SMV/Portal vein. This a major limiting factor.[19]

Conservative resection with aggressive medical management seems to be the order of the day. Laparoscopic approach is best suited for this with studies showing excellent safety and good functional outcome.

Our study is also testimony to the above documented facts.

Limitations

Larger sample size with longer follow-up is required to authoritatively advocate the advantages of laparoscopic pancreatectomy in the management of PHHI.

CONCLUSION

Laparoscopic pancreatectomy is feasible and safe in the management of PHHI. The procedure has a steep-learning curve, especially with regard to the removal of the uncinate process and also the removal of a portion of the head without injuring the common bite duct. Less intraoperative complications such as blood loss, associated morbidity, and reduced duration of hospital stay makes laparoscopic pancreatectomy a potent arrow in a surgeon’s quiver.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.