Evaluation of a Magnetic Compression Anastomosis for Jejunoileal Partial Diversion in Rhesus Macaques

Abstract

Purpose: Metabolic surgery remains underutilized for treating type 2 diabetes, as less invasive alternative interventions with improved risk profiles are needed. We conducted a pilot study to evaluate the feasibility of a novel magnetic compression device to create a patent limited caliber side-to-side jejunoileal partial diversion in a nonhuman primate model.

Materials and methods: Using an established nonhuman primate model of diet-induced insulin resistance, a magnetic compression device was used to create a side-to-side jejunoileal anastomosis. Primary outcomes evaluated feasibility (e.g., device mating and anastomosis patency) and safety (e.g., device-related complications). Secondary outcomes evaluated the device's ability to produce metabolic changes associated with jejunoileal partial diversion (e.g., homeostatic model assessment of insulin resistance [HOMA-IR] and body weight).

Results: Device mating, spontaneous detachment, and excretion occurred in all animals (n = 5). There were no device-related adverse events. Upon completion of the study, ex vivo anastomoses were widely patent with healthy mucosa and no evidence of stricture. At 6 weeks post-device placement, HOMA-IR improved to below baseline values (p < 0.05). Total weight also decreased in a linear fashion (R² = 0.97) with total weight loss at 6 weeks post-device placement of 14.4% (p < 0.05).

Conclusion: The use of this novel magnetic compression device to create a limited caliber side-to-side jejunoileal anastomosis is safe and likely feasible in a nonhuman primate model. The observed glucoregulatory and metabolic effects of a partial jejunoileal bypass with this device warrant further investigation to validate the long-term glucometabolic impact of this approach.

Keywords: Insulin resistance; Magnamosis; Magnetic compression device; Metabolic surgery; Small-bowel anastomosis; Staple-free anastomosis.

Fig. 1

The Magnamosis device ((cc) Magnamosis, Inc. 2021; San Francisco, CA, USA)

Fig. 2

HOMA-IR 3 and 6 weeks after intervention compared to pre-intervention baseline (week 0). Data are reported as median (IQR). *Statistically significant change from baseline (p < 0.05). HOMA-IR, homeostatic model assessment of insulin resistance

Fig. 3

Weight loss from time of intervention (week 0) to 6 weeks. A strong linear decrease (R.² = 0.97) results in a 14.4% fall in total body weight at 6 weeks (p = 0.004). Data are reported as mean (SEM). *Statistically significant change from baseline (p < 0.05)

Fig. 4

Fasting plasma concentrations of total cholesterol (A), triglycerides (B), HDL-c (C), and LDL-c (D) at 3 and 6 weeks after intervention compared to pre-intervention baseline (week 0). Data are reported as median (IQR). *Statistically significant change from baseline (p < 0.05). LDL-c, low-density lipoprotein cholesterol; HLD-c, high-density lipoprotein cholesterol

Fig. 5

Plasma concentrations of adiponectin (A), leptin (B), and the adiponectin/leptin ratio (C) at 3 and 6 weeks after intervention compared to pre-intervention baseline (week 0). Data are reported as median (IQR). *Statistically significant change from baseline (p < 0.05)