We thank Dr. Cersosimo for his letter addressing our editorial and are pleased to address his concerns (1,2). In our article “To Pump or Not to Pump?” (2), we pointed out that both emotions and scientific evidence have contributed to the debate on the role of continuous subcutaneous insulin infusion (CSII) versus multiple daily injections (MDI). It is worth remembering that the only real benefit of CSII relates to the kinetics of delivery of subcutaneous insulin. We emphasized that all previous studies in which NPH insulin was used as the background insulin in MDI were of limited value because insulin glargine now provides a more physiological insulin profile (3). In addition, very few studies comparing CSII and MDI used a randomized, blinded treatment design. Without such a protocol, selection bias is very difficult to exclude. This problem is illustrated by the study cited by Cersosimo, which was neither randomized nor blinded (4). In this study, a significant difference in the two groups was present, both in the starting HbA1c levels (8.0 ± 1.2% for CSII and 8.6 ± 1.0% for MDI) and in the number of individuals in each group (35 for CSII and 50 for MDI). Without equivalence of groups at the initiation of the study, conclusions concerning their outcome parameters are difficult to make.
A second issue that is frequently overlooked when comparing CSII with MDI is the degree of interaction between the volunteer and the investigator’s health care team. Initiating CSII requires extensive training in diet, carbohydrate counting, hypoglycemia awareness, attention to pump detail, and insulin delivery algorithms. This training and close follow-up are often more extensive than that which occurs with initiation of MDI. What percentage of the high success rate reported with CSII therapy is due to intensive educational and motivational training is unknown.
A third issue of concern is the cost of CSII versus MDI. We agree with Cersosimo, from the viewpoint of long-term cost analysis, that CSII may not be more costly than MDI. In fact, it may be cost effective. However, cost can be a major obstacle to the individual diabetic patient who may not have medical insurance and is therefore not be able to cover the immediate costs of CSII from his or her own resources. For many individuals in the U.S. currently without health insurance, starting CSII is not financially possible.
Our editorial attempted to make one major point—that additional data are urgently required before CSII can be broadly recommended over MDI in the general population of people with type 1 diabetes. Not only do we need studies utilizing insulin glargine in the MDI regimen, but these studies also need to be randomized and blinded to the greatest extent possible. Until such data become available, the decision to prescribe either CSII or MDI should be based on the criteria that we listed in our editorial. The health care team must determine which treatment modality has the greatest potential for benefit in each diabetic patient. K.M. Bolderman (5) appropriately states this concept in her excellent book on CSII by stating, “Discovering the character and source of motivation through careful screening of the patient is the key to ensuring success in pump therapy.”
References
Address correspondence to David S. Schade, MD, University of New Mexico School of Medicine, Department of Internal Medicine/Division of Endocrinology and Metabolism, University of New Mexico Health Sciences Center, 5-ACC, Albuquerque, NM 87131. E-mail: dschade@salud.unm.edu.
