In Defense of Insulin Pumps?

I never thought I would write something with a headline like this -- as a former insulin pump wearer myself, I usually argue that the benefits of insulin pumps are too often vastly overstated and that I've attained comparable glycemic control without the expensive device. That's not to say that some people don't benefit from them, particularly people whose basal rates vary considerably throughout the day, or those whose sensitivity to insulin is such that they require dosage precision smaller than 1/2 units, but I still believe that pump evangelists suggest that the answer to everyone's diabetes management can be answered with a pump.

Regardless, at the beginning of every month, when the new editions of various scientific and medical journals are released, my RSS reader is bombarded with news stories. I typically scan the headlines for these, and then promptly discard a vast majority of these stories.

The May issue of the medical journal Pediatrics features an FDA study (which is by itself, unusual, since the FDA has largely been criticized for its failure to examine Adverse Event data involving all of its approved products, we need look no further than last year’s scandal involving the type 2 diabetes treatment Avandia as an example). Nevertheless, this month, a submission entitled "Adolescent Use of Insulin and Patient-Controlled Analgesia Pump Technology: A 10-Year Food and Drug Administration Retrospective Study of Adverse Events", the full text is available free) by Judith U. Cope, et al happened to make the Associated Press (AP) headlines yesterday. The headline read: "Insulin Pumps Linked to Teen Deaths".

Ignoring the headline for a moment, readers should know that the study looked at Adverse Events in the FDA's Manufacturer and User Facility Device Experience Database reported between January 1, 1996 and December 31, 2005. It looked at 1,674 reports, of which 1,594 were specifically pertaining to insulin pumps. Pediatric patients aged 12 to 21 years were the focus.

The AP article and headline implies that there is a huge, perhaps previously undiscovered risk associated with insulin pumps, yet the statistics in the journal article suggest otherwise. The study found 13 deaths reported, and simple math applied here indicates the incidence of actual deaths was 0.81% (in other words, less than 1%), but when compared to Adverse Event reports involving insulin without a pump, the number cannot even be considered statistically significant. Why, then, the headline of "Insulin Pumps Linked to Teen Deaths"? Because that creates controversy, draws readers and hopefully leads to advertising revenue for the newspapers.

Interesting enough, the most significant finding was that almost 62% of the reports involved patient problems of HYPERglycemia, and nearly 47% of those indicated that the patient had DKA. 7.4% of the reports revealed problems of underinfusion, although in truth, the pump may have been blamed for problems with subcutaneous infusion. The device problems included error messages (n=55), incorrect use (n=24), alarm problems (n=29), loosening and/or occlusion of the catheters (n=22), bent cannulas (n=22), and screen display problems (n=14). There were also 31 reports of the device’s need for repair, replacement, or removal and 19 reports of device failure and/or failure to deliver insulin.

Less than 11% of the reports involved HYPOglycemia or overdelivery of insulin. Interestingly, in around 36% of the hypoglycemic reports, the device problem was coded as "unknown" (revealing by itself in that many Adverse Event reports are inconsistently coded), so they may have been linked to patient dosage errors, perhaps when hastily entering a bolus amount, or simply an error in calculating the dose ... again, is this the fault of the pump manufacturer, the patient or simply the overly-complex method of treatment being demanded of pediatric patients? Let me add that these are only those reports that were filed with the FDA, but we know from experience that a vast majority of reports are never recorded for a host of different reasons, most notable are that #1, reporting is strictly voluntary, even among Emergency Rooms and #2, the reporter must have reason to think the Adverse Event is linked to the device rather than patient noncompliance or error, which seldom happens.

It would be even more interesting to see whether these same reports correlated with a similar Adverse Event filing for insulin – my suspicion is that they would not, but somehow those questions are never asked by our regulators!

Back to the Pediatrics analysis on insulin pumps: approximately 80% of the reports resulted in hospitalization. But the top 3 issues associated with these reports were all linked to patient education, noncompliance, and perhaps most troubling, problems during sports or other activities. Just 4 reports suggested risk-taking behaviors (including 2 reports of insulin overbolusing that were thought to be suicide attempts), but we already know that suicide rates tend to jump during the teenaged years for a host of different reasons, but none can be clearly be "blamed" on the insulin pump itself, but the method of insulin delivery, which is subcutaneous delivery of insulin, the very same way insulin is injected with syringes or pens.

The study's author took pains to note that they did not advise against using the devices, but did call for more study to address safety concerns in teens and younger children using pumps. They wrote "This large number of reports may reflect the expanding use in the pediatric age group, but the unique challenges posed by adolescent users should be considered."

The FDA's Dr. Judith Cope, lead author of the analysis told the AP "Parental oversight and involvement are important. Certainly teenagers don't always consider the consequences."

Our own Kelly Close was also quoted in the AP article, although I suspect her quote might have been taken a bit out of context. Her sound bite was reported as follows: "The pumps allow young people to live more normal lives, giving themselves insulin discreetly in public. And they're a growing segment of diabetes care, with $1.3 billion in annual sales worldwide." She said 100,000 teenagers may be using them. The reason I suspect that statement was taken out of context is because it's unlikely that knowing Kelly, I doubt would say that the main benefit of insulin pumps is the be able to give insulin to themselves "discretely in public", most likely she told the interviewer a number of benefits pumping can provide, and this was an easy one to excerpt.

Regardless, the AP article is guilty of reporting only half-truths to get a headline, but the full report beneath the headlines are usually far more informative, as was the case here. However, I would mention that Adverse Event reporting is an area the FDA has done a poor job of gathering information in the past, although the Agency held a workshop on this very subject earlier this year and even sought public commentary on the matter -- I know, because I submitted a lengthy document myself, co-authored with a number of other people. The workshop transcript from the FDA's recent workshop can be found online here, but clearly, the Food and Drug Administration has a l-o-o-o-o-o-o-o-n-g way to go when it comes to gathering information about adverse events. But blaming the pump itself for problems that are not clearly tied to these Adverse Events seems to be irresponsible reporting at its worst.