Implantable pumps, insulin

P. J. Blackshear, Implantable pumps for insulin delivery Current clinical status, in Drug Delivery Systems, Fundamentals and Techniques (P. Johnson and J. G. Lloyd-Jones, eds.), Ellis Horwood, Chichester, 1987, p. 139. 

A new trend in the delivery of medicines is to employ a device component. This may be an implantable pump for insulin, a metallic stent coated with a drug, or unit capable of rapidly vaporizing a discrete dose for inhalation. Such products are regulated by the FDA as "combination" products and may be reviewed by multiple Centers within the Agency, which may require additional levels of documentation to support the product design. 

Repeated episodes of catheter obstruction by fibrin clots or omental encapsulation can be a problem during continuous peritoneal insulin infusion from implanted pumps (SEDA 20, 397). In the encapsulated tissue, collagen fibrosis, inflammatory reactions with lymphocytes, and amyloid-like deposits reacting to anti-insulin antibodies can occur higher macrophage chemotaxis may also promote these processes. 

Diabetes mellitus in a 36-year-old man with acute pancreatitis could not be controlled with continuous subcutaneous insulin infusion, even with doses up to 1800 U/ day, because of insulin resistance (168). Intravenous insulin by pump had to be stopped because of a catheter infection. The continuous subcutaneous infusion of freeze-dried insulin and the addition of aprotinin, a protease inhibitor, soluble dexamethasone or prednisolone, and intravenous immunoglobulin was ineffective. An implantable pump for intraperitoneal delivery established good regulation at a dosage of 30 U/day. 

Continuous intraperitoneal insulin infusion with implantable pumps has been assessed in 34 patients with poorly controlled diabetes (231). In two patients, the pump was explanted in one patient with Werner s syndrome (no subcutaneous fat) the pump was explanted because of infection in the pocket, and one pump was explanted because the patient had local complaints and psychological problems. One patient refused to be included. Patients were followed for 58 months. HbAic fell from 10.0 to 9.0% in the first year and remained there. Median days in hospital fell from 45 to 13 after 1 year. The quality of life was relatively low and many had psychiatric problems. Although long-term glycemic control improved and lengths of hospital stay were reduced, normal glucose control and normal quality of life could not be achieved. 

Problems with insulin delivery in implanted pumps are difficult to correct. A change in Hoechst 21 pH-neutral semisynthetic insulin 400 U/ml in accordance with regulations of the European Pharmacopoeia (SEDA-20, 397) resulted in more frequent clogging when this insulin was used in the Minimed 2001 implantable pump (MIP 2001). From October 1995 to October 1996, 17 pumps were implanted (241). The refilling period was reduced from 90 to 30-45 days and the reservoirs were washed with insulin-free buffer before each refill. Backflow was seen in 13 pumps after a mean period of 7.2 months. Modification of the manufacturing process produced 21PH ETP insulin (human semisynthetic insulin, Genapol-stabilized) 400 U/ml, Hoechst, with improved stability since July 1997. All pumps were specifically cleaned before the new insulin was used for refill. The refill period was increased from 38 to 78 days. In 16 pumps, only one backflow was seen after 14 months. 

The incidence of catheter blockage did not change. The better stability of this insulin for implantable pumps has been confirmed in a study in which 88 pumps were refilled every 45 days and 108 pumps every 90 days (242). 

Riveline JP, Capeau J, Robert JJ, Varroud-Vial M, Cerf-Baron I, Deburge A, Charpentier G. Extreme subcutaneous insulin resistance successfully treated by an implantable pump. Diabetes Care 2001 24(12) 2155-6. 

Pinget M, Jeandidier N. Long term safety and efficacy of intraperitoneal insulin infusion by means of implantable pumps. Horm Metab Res 1998 30(8) 475-86. 

Kessler L, Tritschler S, Bohbot A, Sigrist S, Karsten V, Boivin S, Dufour P, Belcourt A, Pinget M. Macrophage activation in type 1 diabetic patients with catheter obstruction during peritoneal insulin delivery with an implantable pump. Diabetes Care 2001 24(2) 302-7. 

Renard E, Souche C, Jacques-Apostol D, Lauton D, Gibert-Boulet F, Costalat G, Bringer J, Jaffiol C. Improved stability of insulin delivery from implanted pumps using a new preparation process for infused insulin. Diabetes Care 1999 22(8) 1371-2. 

In the MiniMed implantable pump, a piston pump drives insulin through the delivery catheter. A patented solenoid motor controls the piston movement, to aspirate insulin from the reservoir chamber into the piston chamber and then push it through the insulin delivery catheter. 

In addition to needles and syo inges, alternative techniques for insulin administration have been developed, some availing themselves of the kinetics of insulin insulin pens (supplied preloaded or with replaceable cartridges), external infusions and implantable pumps. These latter are convenient for cm accurately controlled continuously functioning biofeedback system, but pose difficulties for routine replacement in insulin deficiency. Therefore sustained-release (depot) formulations are used to provide an approach reasonably near to natural function and compatible with the convenience of daily living. An even closer approach is provided by the development of (at present inevitably expensive) miniaturised infusion pumps which can be used by reliable patients. 

Drug delivery is covered more adequately, with a variety of MeSH terms including drug delivery systems, drug carriers, and infusion pumps, which has narrower terms such as infusion pumps, implantable, and insulin infusion systems. 

Pumps are surgically implanted subcutaneously, normally in the abdomen, and the catheter tip is usually placed in the peritoneal space, as intravenous deUvery generally has been less successful. The catheter is undoubtedly the weakest point in implantable pumps, and the most significant complication with these pumps has been recurrent episodes of apparent pump slowdown related to catheter obstruction. In the early days, these events were most often attributed to insulin precipitation in the delivery cannulas. However, more carefiil inspection revealed that the problem in the majority of cases was, in fact, caused by small fibrin plugs at the end of the catheter or even macroscopic tissue encapsulation of the catheter tip. 

In a recent study, it was found that treatment of diabetic patients by using an implantable pump to dehver insulin into the peritoneal cavity leads to a greater immunogenicity of insulin as compared to conventional therapy (Lassmann-Vagueeta/., 1995). 

Because an implanted pump contains thousands of units of insulin, safety considerations are of vital importance. Therefore, such a device must be extremely rehable and safe, and fijrther long-term assessment of these devices is needed. At present, implanted pumps are considered research devices, which are only available as part of approved protocols (Saudek, 1993). The financial costs and need for recurrent surgery makes it unlikely that implantable insulin pumps ever will be a widely used treatment option. 

---