Diabetes Care, Vol 16, Issue 8 1103-1115, Copyright © 1993 by American Diabetes Association
A highly successful and novel model for treatment of chronic painful diabetic peripheral neuropathy
MA Pfeifer, DR Ross, JP Schrage, DA Gelber, MP Schumer, GM Crain, SJ Markwell and S Jung
Diabetes Research and Treatment Center, Southern Illinois University, Springfield 61702.
OBJECTIVE--To investigate why, in spite of a vast variety of treatment
agents, the alleviation of pain in patients with diabetic neuropathy is
difficult. Previous studies have not used a treatment algorithm based on
anatomic site and neuropathophysiological source of the neuropathic pain.
RESEARCH DESIGN AND METHODS--A model that categorizes the types of pain
into three groups (superficial, deep, and muscular) was applied in 75
diabetic patients with chronic (> 12 mo) painful distal symmetrical
polyneuropathy in a controlled case series. Twenty-two patients were
untreated and 53 patients were treated with imipramine +/- mexiletine for
deep pain, capsaicin for superficial pain, and stretching exercises and
metaxalone +/- piroxican for muscular pain. Each type of pain was scored
separately on a scale of 0 (none) to 19 (worst), and the total of all three
types was used as an index of overall pain. Ability to sleep through the
night was scored by a scale of 1 (never) to 5 (always). RESULTS--No
significant differences were observed in initial pain scores, sleep scores,
demographics, biochemistries, or physical findings between the two groups.
After 3 mo a significant improvement in scores was noted in the treated but
not the untreated patients. In addition, a significant difference was found
in the change of scores between the treated and untreated patients: total
pain (-18 +/- 2 vs. 0 +/- 2), deep pain (-7 +/- 1 vs. 0 +/- 1), superficial
pain (-5 +/- 1 vs. 0 +/- 1), muscular pain (-6 +/- 1 vs. 0 +/- 1), and
sleep (1.2 +/- 0.2 vs. 0.2 +/- 0.2), all P < 0.0001. In treated patients
21% became pain-free (total pain < 2), 66% had improvement (decrease in
total pain > 5, but not total elimination of painful symptoms), and 13%
were considered treatment failures (a decrease in total pain of < or =
5). This compares with 0 (P < 0.02), 10 (P < 0.0001), and 90% (P <
0.0001), respectively, in the untreated patients. CONCLUSIONS--This study
presents a new rationale and hypothesis for the successful treatment of
chronic painful diabetic peripheral neuropathy. It uniquely bases the
treatment algorithm on the types and sources of the pain.