Increased Restenosis in Diabetes Mellitus After Coronary Interventions Is Due to Exaggerated Intimal Hyperplasia
A Serial Intravascular Ultrasound Study
Background The increased risk of restenosis after catheter-based coronary interventions in diabetic patients has not been determined. Intravascular ultrasound (IVUS) has shown that the decrease in arterial area is responsible for most of the late lumen loss in nonstented lesions and that intimal hyperplasia is responsible for all of the late lumen loss in stented lesions.
Methods and Results Serial (postintervention and follow-up at 5.6±3.3 months) IVUS was used to study 251 native coronary lesions in 241 patients; 63 patients had treated diabetes mellitus (oral hypoglycemic drugs or insulin). Interventional procedures included percutaneous transluminal coronary angioplasty, directional or rotational atherectomy, excimer laser angioplasty, or Palmaz-Schatz stents. The external elastic membrane (EEM), stent, and lumen areas were measured. The plaque+media (P+M) area in nonstented lesions was calculated as EEM minus lumen area, and the intimal hyperplasia (IH) area in stented lesions was calculated as stent minus lumen area. The anatomic slice selected for serial analysis had an axial location within the target lesion at the smallest follow-up lumen area. Nonstented lesions in diabetics and nondiabetics had a similar decrease in EEM cross-sectional area (CSA; 1.9±2.8 versus 1.8±4.2 mm2; P=.6350). However, nonstented lesions in diabetics had a greater increase in P=M CSA (1.3±2.8 versus 0.6±2.5 mm2, P=.0720), and the increase in P=M CSA contributed a greater percentage to the decrease in lumen CSA. In stented lesions, the decrease in lumen CSA (5.2±2.5 versus 2.0±2.3 mm2) and the increase in IH CSA (5.0±2.8 versus 1.8±2.0 mm2) were greater in diabetics than nondiabetics (P=.0009 and P=.0007, respectively). These findings were even more striking in (nonstented and stented) restenotic lesions.
Conclusions Serial IVUS analysis showed that the main reason for increased restenosis in diabetes mellitus was exaggerated intimal hyperplasia in both stented and nonstented lesions.
Restenosis remains a major limitation to percutaneous coronary revascularization. Patients with diabetes mellitus have a greater incidence of restenosis, which is associated with increased late morbidity and mortality.1 2 Animal models, human necropsy studies, and analyses of retrieved atherectomy specimens originally suggested that an exaggeration of the normal reparative processes after angioplasty-induced local vessel trauma leads to uncontrolled smooth muscle cell proliferation and restenosis. Conversely, recent serial IVUS studies showed that arterial remodeling (the chronic decrease in arterial CSA) was responsible for most of the late lumen loss in nonstented lesions and that in-stent neointimal hyperplasia was responsible for all of the late lumen loss in stented lesions.3 4 The present study used IVUS to study the reason for exaggerated restenosis in diabetic patients.
From July 1991 to April 1996, serial (postintervention and follow-up at 5.6±3.3 months) IVUS and QCA were used to study 251 native coronary lesions in 241 patients. This was an inclusive series of patients studied after intervention and at follow-up for recurrent symptoms or as part of clinical protocols.
There were 149 men and 92 women (mean age, 58±11 years). On the basis of blinded independent chart review, 63 patients had treated diabetes mellitus (insulin or oral hypoglycemic drugs). Lesion location was left main (n=6), left anterior descending (n=104), left circumflex (n=37), and right coronary (n=104) arteries. Interventional procedures included PTCA (n=35), directional (n=111), or rotational (n=41) atherectomy; excimer laser angioplasty (n=24); or Palmaz-Schatz stents (n=40). Lesion location and patterns of device used were similar in diabetic and nondiabetic patients.
All films were reviewed by individuals at a core angiographic laboratory who were blinded to the diabetes classification and IVUS results. With the use of an automated edge detection algorithm (ARTREK, Quantitative Cardiac Systems) and the outer diameter of contrast-filled catheters for calibration, MLD, reference diameter, and percent diameter stenosis were measured from multiple projections; the results in the “worst” view were recorded. Angiographic restenosis was defined as a follow-up percent diameter stenosis of ≥50%.
IVUS studies were performed using one of two systems. The first (InterTherapy Inc) incorporated a single-element 25-MHz transducer and an angled mirror mounted on the tip of a flexible shaft within a 3.9F short monorail imaging sheath. The second (CardioVascular Imaging Systems, Inc) used a single-element beveled transducer mounted on the tip of a flexible shaft within a 3.2F short monorail imaging sheath. With both systems, the transducer was rotated at 1800 rpm and withdrawn automatically at 0.5 mm/s. After intervention and at follow-up, 0.2 mg nitroglycerin IC was given, the IVUS catheter was advanced 5 to 10 mm distal to the lesion, and a complete imaging run was performed from beyond the lesion to the aorto-ostial junction with the motorized transducer pullback device. Studies were recorded on 1/2-in high-resolution super VHS tapes for off-line analysis.
Quantitative IVUS Analysis
With computerized planimetry, quantitative image analysis was performed by a single individual blinded to the QCA results and diabetes classification. By use of one or more reproducible axial landmarks (eg, aorto-ostial junction, side branches, or unusually shaped calcium deposits) and a known pullback speed, identical image slices could be identified for serial analysis. The image slice analyzed had an axial location within the target lesion at the smallest follow-up lumen CSA (rather than at the smallest postintervention lumen CSA). In practice, the follow-up study was analyzed first to identify the image slice with the smallest lumen; then, the distance from this image slice to the closest identifiable axial landmark was measured using seconds of videotape; finally, this distance was used to identify the corresponding image slice on the postintervention study. Vascular and perivascular markings were used to confirm image slice identification. If necessary, serial studies were analyzed side by side and imaging runs were studied frame by frame to ensure that the same image slice was measured.
Validation of cross-sectional measurements by IVUS and their use and reproducibility in assessing mechanisms of restenosis have been reported previously.3 4 5 6 In nonstented lesions, the EEM and lumen CSA were measured; P+M CSA was calculated as EEM minus lumen CSA. In stented lesions, stent and lumen CSA were measured. IH CSA was defined as stent minus lumen CSA. The EEM CSA (representing the area within the media-adventitia border) was a measure of total arterial CSA. Because IVUS cannot measure media thickness accurately, P+M CSA was used as a measure of plaque. When plaque encompassed the catheter, the lumen was assumed to be the physical, not acoustical, size of the imaging catheter.
Statistical analysis was performed with Statview 4.02 (Abacus Concepts). Quantitative data are presented as mean±SD. Qualitative data are presented as frequencies. Categorical variables were compared by use of χ2 statistics and Fisher's exact test. Continuous variables were compared by use of paired and unpaired Student's t test.
Serial QCA Results
The reference measured 2.81±0.51 mm. After intervention, the MLD increased from 0.91±0.48 to 2.45±0.59 mm (P<.0001); the percent diameter stenosis decreased from 68±16% to 16±14% (P<.0001). At follow-up, the MLD decreased to 1.48±0.89 mm; the percent diameter stenosis increased to 48±27% (both P<.0001 vs postintervention values); and 129 lesions (51%) were restenotic.
Serial IVUS Results in Nonstented Lesions
In nondiabetics, 78% of the decrease in lumen CSA (from 6.8±2.5 mm2 after intervention to 4.5±4.4 mm2 at follow-up, P<.0001) was the result of a decrease in EEM CSA (from 20.0±6.3 to 18.2±6.7 mm2, P<.0001); 22% was the result of an increase in P+M CSA (from 13.2±5.6 to 13.8±5.5 mm2, P=.0090). In diabetics, 59% of the decrease in lumen CSA (from 6.3±2.6 mm2 after intervention to 3.1±2.5 mm2 at follow-up, P<.0001) was the result of a decrease in EEM CSA (from 20.6±6.9 to 18.7±6.3 mm2, P<.0001); 41% was the result of an increase in P+M CSA (from 14.3±5.6 to 15.6±6.0 mm2, P=.0005). Diabetics and nondiabetics had a similar decrease in EEM CSA (see the Table⇓). However, treated diabetics had a greater increase in P+M CSA, and the increase in P+M CSA contributed a greater percentage to the decrease in lumen CSA. These findings were even more significant in restenotic lesions (see the Table⇓ and Figs 1⇓ and 2⇓).
Serial IVUS Results in Stented Lesions
All the decrease in lumen CSA at follow-up (from 8.0±2.4 to 2.8±2.7 mm2 in diabetics and from 7.3±2.9 to 5.3±3.2 mm2 in nondiabetics, both P<.0001) was the result of IH CSA (5.1±2.8 mm2 in diabetics and 2.1±2.0 mm2 in nondiabetics). The decrease in lumen CSA and the increase in IH CSA were greater in diabetics; as in nonstented lesions, these findings were even more striking in restenotic lesions (Table⇑).
Diabetes is a consistent clinical predictor of restenosis after angioplasty and coronary stenting.1 2 7 8 Several studies have reported restenosis rates of 47% to 71% in diabetics.1 2 7 A recent analysis from the Bypass Angioplasty Revascularization Investigation trial8 showed worse 5-year survival rates in diabetics with multivessel disease treated by balloon angioplasty compared with those undergoing bypass surgery. Using QCA, Carrozza et al9 found that diabetics had a greater incidence of restenosis after stenting; this was attributed to greater in-stent IH.
With serial IVUS analysis, the current study found that in nonstented lesions, the decrease in EEM CSA was similar in diabetics and nondiabetics. However, the present study also found that in diabetics exaggerated tissue proliferation was superimposed on the decrease in EEM CSA in nonstented lesions. In diabetics, there was also exaggerated tissue proliferation in stented lesions. These findings were even more striking in restenotic lesions, whether stented or nonstented. Diabetes is associated with hormonal and vascular abnormalities that promote smooth muscle cell proliferation after vascular injury, including injury from catheter-based interventions.10 Increased smooth muscle proliferation in diabetics may result from mitogens (such as platelet-derived growth factor and insulinlike growth factor) that stimulate cell growth and deleterious vascular effects of endothelial dysfunction and excessive extracellular matrix production.10 11 12 13 14
First, because this is a study of patients presenting for follow-up largely for symptomatic recurrence, it may represent a skewed population because of the nature of the follow-up. However, there was no bias toward performing follow-up studies in diabetic versus nondiabetic patients. Second, this study depended on accurate identification of the same anatomic section on serial studies (image slice with the smallest follow-up minimum lumen CSA). This precluded blinded IVUS analysis, may not have accurately reflected the serial changes in minimum lumen CSA, and potentially exaggerated the degree of both acute lumen CSA gain and loss. Third, differences in vascular tone could have contributed to change in lumen and EEM CSA. However, this should not have affected ΔP+M CSA. Fourth, serial IVUS analysis can measure only net changes in P+M CSA. It cannot isolate the relative contributions of atherosclerosis progression/regression, cellular proliferation, matrix deposition, or plaque stabilization to the overall change in P+M CSA. Fifth, a heterogeneous diabetic population was studied; the number of insulin-dependent patients was too small for subset analysis.
In diabetic and nondiabetic patients, there was a similar decrease in EEM CSA; this contributed importantly to late lumen loss in nonstented lesions, especially in nondiabetics. In diabetic patients, however, there was exaggerated tissue proliferation, especially in restenotic lesions. This was seen in both stented and nonstented lesions and may explain the increased rate of restenosis in diabetes mellitus.
Selected Abbreviations and Acronyms
|EEM||=||external elastic membrane|
|MLD||=||minimal lumen diameter|
|P+M||=||plaque plus media|
|PTCA||=||percutaneous transluminal coronary angioplasty|
|QCA||=||quantitative coronary angiography|
This study was supported in part by the Cardiology Research Foundation, Washington, DC.
- Received November 21, 1996.
- Revision received January 22, 1997.
- Accepted January 23, 1997.
- Copyright © 1997 by American Heart Association
Stein B, Weintraub WS, Gebhart SSP, Cohen-Bernstein CL, Grosswald R, Liberman HA, Douglas JS, Morris DC, King SB III. Influence of diabetes mellitus on early and late outcome after percutaneous transmural coronary angioplasty. Circulation. 1995;91:979-989.
Kip KE, Faxon DP, Detre KM, Yeh W, Kelsey SF, Curier JW. Coronary angioplasty in diabetic patients: the National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angioplasty Registry. Circulation. 1996;94:1818-1825.
Mintz GS, Popma JJ, Pichard AD, Kent KM, Satler LF, Wong C, Hong MK, Kovach JA, Leon MB. Arterial remodeling after coronary angioplasty: a serial intravascular ultrasound study. Circulation. 1996;94:35-43.
Hoffmann R, Mintz GS, Dussaillant GR, Popma JJ, Pichard AD, Satler LF, Kent KM, Griffin J, Leon MB. Patterns and mechanisms of instent restenosis: a serial intravascular ultrasound study. Circulation. 1996;94:1247-1254.
Stout RW, Bierman EL, Ross R. Effect of insulin on the proliferation of cultured primate arterial smooth muscle cell. Circ Res. 1975;36:319-327.
Bornfeldt KE, Raines EW, Nakano T, Graves TN, Krebs EG, Ross R. Insulin-like growth factor-1 and platelet derived growth factor-BB induce direct migration of human smooth muscle cells via signaling pathways that are distinct from those of proliferation. J Clin Invest. 1994;93:1266-1274.
Kirstein M, Brett J, Radoff S, Ogawa S, Stern D, Vlassara H. Advanced protein glycosylation induces selective transendothelial human monocyte chemotaxis and secretion of PDGF: role in vascular disease in diabetes and aging. Proc Natl Acad Sci U S A. 1990;87:9010-9014.