Nailfold
Videomicroscopy and Local Cold in Type I Diabetics P.
Test
Gasser, M.D., F.I.C.A. and W.
Berger, M.D.*
BASEL, SWITZERLAND
Abstract
Twenty-five type I diabetics (9 men, 16 women) with a mean disease duration of 19.9 years and 25 sex- and age-matched healthy control subjects were studied.
Digital capillary blood flow measurements in combination with a local cooling test were assessed by nailfold videocapillaroscopy using the technique of flying spot and compared with nailfold capillary microscopy in four subgroups divided according to (A) disease duration, (B) retinopathy, (C) fasting blood 1c values. Differences in morphologic capillary diameglucose level, and (D) HbA ters and capillary density were found between the diabetics and the healthy control subjects. These were attributable to the patients with retinopathy. The hemodynamic findings at rest and after local cooling were unable to differentiate either between type I diabetics and healthy controls or within the different subgroups. Introduction Abnormalities of microcirculation of skin blood flow are known to occur in diabetes mellitus,’ but the exact mechanism by which diabetic microangiopathies occur are up to now unclear. Altered metabolism of glycoprotein,2 hemorrheologic changes with increased blood viscosity,’ and platelet dysfunction4 have been described. In view of these hypotheses it is thought that the genesis of microangiopathies of diabetes may be a multifactorial process. Discrepancies between the total circulation of the skin area and the nutritional status of the tissue can be seen, especially in patients with diabetes.’ Nailfold capillary microscopy is a noninvasive method to investigate the nutritional status of digital microcirculation. From the Department of Internal Medicine, St. Claraspital, Basel and *the Department of Internal Medicine, University of Basel, Basel, Switzerland This study was supported by the Swiss National Science Foundation, Grant No. 32-9367.87.
395
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
396 The aim of the present study was to assess the hemodynamic and the morphologic microvascular patterns in combination with a local cooling test in type I diabetes by nailfold capilla-
roscopy. Patients and Methods
Twenty-five insulin-dependent type I diabetics (9 men, 16 women) who regularly attend a diabetic outpatients clinic were studied. Their details are summarized in Table I. Each diabetic patient underwent clinical assessment including ophthalmoscopy. Glycosylated hemoglobin (HbA,~) and fasting blood glucose levels were measured in all the diabetic patients. Their mean age was 46.4 years (range twenty-six to sixty-eight). The mean duration of diabetes was 19.9 years (range eight to forty), mean HbA,, 7.0% (range 5.5-9.4), and mean fasting blood glucose level 7.9 mmoll-’ (range 5.0-13.1). Their mean systolic blood pressure was 125 (± 17) mmHg, diastolic pressure 80 (± 10) mmHg. All diabetics had absence of sticktest-positive proteinuria and none had clinically evident neuropathy. Thirteen of them presented a background retinopathy as judged by independent ophthalmologic review in the previous twelve months. All patients had absence of changes in diabetic therapy during the previous four weeks. Patients with anemia, hypertension, or renal disease, and those receiving vasoactive drugs were excluded from the study. TABLE I Clinical Data of Patients
A group of healthy nondiabetic volunteers of similar age
(46.2 years; range twenty-six to distribution 16 acted as control subjects. They were re(9 men, sixty-five) women) cruited from hospital staff and volunteers. Their mean systolic blood pressure was 121 (± 15) mmHg, mean diastolic pressure 79 (±9) mmHg. None of the subjects had symptoms of peripheral arterial disease, Raynaud’s phenomenon, or anemia or were receiving any vasoactive and
sex
drugs. In order to look at the influence of disease duration, fasting blood glucose and HbA,, levels, and background retinopathy, the morphologic and hemodynamic parameters in nailfold capillaries within type I diabetics were compared in the following subgroups: Group A: Diabetics with disease duration of more than (n=14) or less than or equal to (n =11 ) fifteen years. Group B: Diabetics with background retinopathy (n=13) or without retinopathy (n=12).
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
397
Group C:
Group
D:
Diabetics with fasting blood glucose level more than (n=10) or less than or equal to (n =15) 8 mmoll-’ . Diabetics with at least two HbAlc values more than (n =11 ) or less than or equal to
(n =14) 7 .0 % . The data of each diabetic
subgroup
were
additionally compared
with those of the
healthy
control group.
Glycosylated hemoglobin was measured by the technique of high-pressure liquid chromatography. The normal range in our laboratory is 3 . 5-5 .5 % . Plasma glucose was measured by a glucose oxydase method. Patients were studied in the morning after having taken their normal breakfast and insulin dose. The finger nailfold capillaries were studied with the help of an incident-light microscope attached to a television monitor.6 The diameters of the nailfold capillaries were measured by means of the 10 x objective before hemodynamic evaluation was begun. The density of the capillary loops represents the number of capillaries per millimeter in the first row of loops parallel to the epidermis. To measure the capillary blood cell velocity (CBV), the television
pictures were videotaped and analyzed during playback by the flying spot technique as previously described.’ Studies were performed following acclimatization during thirty minutes in a room maintained at constant temperature (23°C). CBV was measured at each session five times.’ The first measurement (baseline) was done under normal natural fingertip temperature (V 1 ) . The second reading was performed after the fingertip had been immersed in a warm water bath (40 ° C) for three minutes (V2). The third measurement was done after the observed skin area had been cooled for sixty seconds by blowing rapidly decompressed carbon dioxide of about -15°C over the nailfold (V3). The fourth reading was done after one minute (V4), and the fifth measurement, finally, after two minutes of spontaneous recovery (V5). In cases where the blood flow ceased during local cooling, the duration of the blood flow standstill was measured in seconds. Skin temperature in the investigated area was measured with a thermocouple. Data Analysis Results have been expressed as the mean and the standard deviation. To examine for statistically significant differences between control, study patients, and subgroups, the Mann-Whitney U test (two-tailed) was used. Frequency distribution of nailfold capillary density was determined by contingency tables (Chi square). The Kendall’s rank correlation coefficient was 7
used to determine correlations.
Results Pattern values for the diameters of the arteriolar and venular limbs of the capillary loops (p = 0.01 ) , as well as the capillary width (p = 0.02), showed that these were significantly larger in type I diabetics than in healthy controls (p < 0.05; Table II). The frequency distribution for capillary density showed a significant reduction (p = 0.0001 ) in diabetic patients as compared with healthy control subjects (Table II). Capillary density did not correlate with age, duration of diabetes, HbA,~, or retinopathy. Retinopathy, on the other hand, correlated with the duration of the disease (tau = 0 . 3 8 ; p = 0 . 020) .
Morphology The
mean
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
398 Morphologic Findings
*Number of nailfold
capillary loops
on
TABLE II in Type I Diabetics
Nailfold Capillaries
(Values Are MeanstSD)
per millimeter
Blood Cell Velocity CBV at rest (V 1 ) and after local cooling (V3) were not significantly different from healthy controls or from the values in the different subgroups (Table III). An examination of the mean values of the blood velocities after warm water bath (V2) and in the spontaneous recovery (V4, V5) revealed no conspicuous differences. We found a flow stop with cold exposure in 5/25 patients with type I diabetes (mean duration 9.1 ± 18 . 5 s) and in 1/25 normal controls (12 s). Flow stop in diabetics was not correlated with a lower skin temperature or a history of cold hands.
Capillary
TABLE III Skin
Temperature (°CJ and Capillary Blood Flow Velocity Before (Vi) and After Cold Test (V3) in Type I Diabetics (Values Are Means f SD)
Skin Temperature There was no significant difference in skin temperature between healthy controls and all type I diabetic patients. However, patients with retinopathy tended to have lower skin temperature (p=0.05) than diabetics without retinopathy (Table III). Discussion In our study we found that the nailfold capillaries of type I diabetics differed from those of normal subjects both in an enlargement of the arterial and venular limb and in a significant reduction of the capillary loop density. Retinopathy as major manifestation of the microvascular disease in diabetes mellitus did not correlate with capillary loop density. Our findings of
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
399 abnormalities in diabetics are in agreement with those of Bollinger et a1,9 who fluorescence used videomicroscopy, as well as with those of Fagrell et al,5 Lagrue et al,’° and Tubiana-Rufi et al,&dquo; who found morphologic irregularities by vital nailfold microscopy in such
morphologic patients.
Within type I diabetics we were not able to distinguish between the gross capillary loop morphology of the different subgroups. This may be due to the small number of patients. The comparison of the diabetic subgroups with the healthy control subjects suggests that the enlargement of the morphologic parameters is attributable to the patients with retinopathy, because the mean values of the morphologic parameters of the healthy nondiabetics and the diabetics without retinopathy are nearly equal. Our study has not identified any statistical differences in CBV by nailfold videomicroscopy between diabetic patients and normal control subjects either at rest or after local cooling. In addition, our analyses showed no significant changes of CBV at rest and after local cold exposure in relation to the different fasting blood glucose levels or HbA,~ values. It must be considered, however, that our small group of diabetic patients is rather homogeneous and, except for the patients with retinopathy, nearly free of microangiopathy. Similar velocities at rest have been obtained by Tooke et al’2 in diabetic patients with various disease durations and levels of diabetic control. Some authors suggest that the high skin blood flow observed during poor diabetes control is mainly determined by an increase in arteriovenous shunt flow. 13.14 On the other hand, some authors have found differences in the functional behavior of skin microcirculation when measuring reactive hyperemia following the release of one minute’s digital arterial occlusion or when modifying skin blood flow by venous occlusion.&dquo; These studies were, however, performed with the technique of laser Doppler flowmetry, which measures subpapillary plexus and arteriovenous shunt flow in addition to nutritive capillary flow.’6 Consequently it is not surprising that the results for CBV did not coincide with those from laser Doppler flowmetry, particularly in diabetics, in whom of peripheral blood flow in favor of nonnutritive shunts may be a feature of the partitioning 13.14 disease. Conclusion It may be concluded that the noninvasive technique of vital nailfold capillaroscopy has allowed a differentiation of morphologic parameters between type I diabetics and normal control subjects. In contrast to retinal fluorescein angiography and fluorescein videomicroscopy, the method of nailfold capillaroscopy without dye is simple and noninvasive. However, in spite of these advantages, the concept that the skin microvascular response to a brief period of local cooling may be a useful test for distinguishing complicated from uncomplicated diabetics, as well as from normal controls, could not be confirmed in this study, where the data base may have been too small and rather homogeneous.
Acknowledgments This
study was supported by the Swiss National Science Foundation, grant no. 32.9367.87.
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
400 The authors thank Mrs. B. Dubler for her excellent technical assistance and Mr. S. Christeller, Hoffmann-La Roche, Basel, for statistical advice and for making helpful suggestions.
Paul Gasser, M.D., F.I. C. A. Department of Internal Medicine St. Claraspital Kleinriehenstrasse 30 CH-4016 Basel, Switzerland
References 1. McMillan DE: The microcirculation: Changes in diabetes mellitus. Mayo Clin Proc 63:517-520, 1988. 2. Spiro RG: Glycoproteins and diabetic micro-angiopathy. In: Marble A, White P, Bradley RF, et al, eds. Joslin’s Diabetes Mellitus. Philadelphia: Lee & Febiger, 1971, p 146. 3. Barnes AJ: Blood viscosity in diabetes mellitus. In: Lowe GDO, Barbanel JC, Forbes CD, eds. Clinical Aspects of Blood Viscosity and Red Cell Deformability. New York: Springer Verlag, 1981, p 151. 4. Chitre AP, Velaskar DS: Role of platelets in diabetic microangiopathy—an additional factor. Angiology
5.
5:458-465, 1989. Fagrell B, Hermansson IL, Karlander SG,
11.
12.
et al: Vital
capillary microscopy for assessment of skin viability and microangiopathy in patients with diabetes mellitus. Acta Med Scand 687:25-28, 1984. Bollinger A, Fagrell B: Clinical Capillaroscopy. Toronto : Hogrefe-Huber, 1990, pp 1-30. 7. Boss Ch, Schneuwly P, Mahler F: Evaluation and clinical application of the flying spot method in nailfold capillary TV-microscopy. Int J Microcirc: Clin Exp 6:15-23, 1987. 8. Gasser P: Capillary blood cell velocity in finger nailfold: Characteristics and reproducibility of the local cold response. Microvasc Res 40:29-35, 1990. 9. Bollinger A, Frey J, Jäger K, et al: Patterns of diffusion through skin capillaries in patients with long-term 6.
diabetes. N 10.
13.
14.
15.
16.
Engl J
Med
307:1305-1310, 1982.
Lagrue G, Hamon P, Maurel A, et conjonctivales et périunguéales dans
al: Angioscopies les diabètes noninsulino-dépendants. Valeurs informationelles des anomalies observées. J Mal Vasc 13:101-105, 1988. Tubiana-Rufi N, Priollet P, Lévy-Marchal C, et al: Detection by nailfold capillary microscopy of early morphologic capillary changes in children with insulin dependent diabetes mellitus. Diabète & Métabolism Paris 15:118-122, 1989. Tooke JE, Lins P-E, Ostergren J, et al: Skin microvascular autoregulatory responses in Type I diabetes: The influence of duration and control. Int J Microcirc: Clin Exp 4:249-256, 1985. Boulton AJM, Scarpello JHB, Ward JD: Venous oxygenation in the diabetic neuropathy foot: Evidence of arteriovenous shunting. Diabetologia 22:6-8, 1982. Edmonds ME, Roberts VC, Watkins PJ: Blood flow in the diabetic neuropathic foot. Diabetologia 22:9-15, 1982. Tooke JE, Ostergren J, Lins P-E, et al: Skin microvascular blood flow control in long duration diabetics with and without complications. Diabetes Res 5:189-192, 1987. Tooke JE, Ostergren J, Fagrell B: Synchronous assessment of human skin microcirculation by laser Doppler flowmetry and dynamic capillaroscopy. Int J Microcirc : Clin Exp 2:277-284, 1983.
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
Videomicroscopy and Local Cold in Type I Diabetics P.
Test
Gasser, M.D., F.I.C.A. and W.
Berger, M.D.*
BASEL, SWITZERLAND
Abstract
Twenty-five type I diabetics (9 men, 16 women) with a mean disease duration of 19.9 years and 25 sex- and age-matched healthy control subjects were studied.
Digital capillary blood flow measurements in combination with a local cooling test were assessed by nailfold videocapillaroscopy using the technique of flying spot and compared with nailfold capillary microscopy in four subgroups divided according to (A) disease duration, (B) retinopathy, (C) fasting blood 1c values. Differences in morphologic capillary diameglucose level, and (D) HbA ters and capillary density were found between the diabetics and the healthy control subjects. These were attributable to the patients with retinopathy. The hemodynamic findings at rest and after local cooling were unable to differentiate either between type I diabetics and healthy controls or within the different subgroups. Introduction Abnormalities of microcirculation of skin blood flow are known to occur in diabetes mellitus,’ but the exact mechanism by which diabetic microangiopathies occur are up to now unclear. Altered metabolism of glycoprotein,2 hemorrheologic changes with increased blood viscosity,’ and platelet dysfunction4 have been described. In view of these hypotheses it is thought that the genesis of microangiopathies of diabetes may be a multifactorial process. Discrepancies between the total circulation of the skin area and the nutritional status of the tissue can be seen, especially in patients with diabetes.’ Nailfold capillary microscopy is a noninvasive method to investigate the nutritional status of digital microcirculation. From the Department of Internal Medicine, St. Claraspital, Basel and *the Department of Internal Medicine, University of Basel, Basel, Switzerland This study was supported by the Swiss National Science Foundation, Grant No. 32-9367.87.
395
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
396 The aim of the present study was to assess the hemodynamic and the morphologic microvascular patterns in combination with a local cooling test in type I diabetes by nailfold capilla-
roscopy. Patients and Methods
Twenty-five insulin-dependent type I diabetics (9 men, 16 women) who regularly attend a diabetic outpatients clinic were studied. Their details are summarized in Table I. Each diabetic patient underwent clinical assessment including ophthalmoscopy. Glycosylated hemoglobin (HbA,~) and fasting blood glucose levels were measured in all the diabetic patients. Their mean age was 46.4 years (range twenty-six to sixty-eight). The mean duration of diabetes was 19.9 years (range eight to forty), mean HbA,, 7.0% (range 5.5-9.4), and mean fasting blood glucose level 7.9 mmoll-’ (range 5.0-13.1). Their mean systolic blood pressure was 125 (± 17) mmHg, diastolic pressure 80 (± 10) mmHg. All diabetics had absence of sticktest-positive proteinuria and none had clinically evident neuropathy. Thirteen of them presented a background retinopathy as judged by independent ophthalmologic review in the previous twelve months. All patients had absence of changes in diabetic therapy during the previous four weeks. Patients with anemia, hypertension, or renal disease, and those receiving vasoactive drugs were excluded from the study. TABLE I Clinical Data of Patients
A group of healthy nondiabetic volunteers of similar age
(46.2 years; range twenty-six to distribution 16 acted as control subjects. They were re(9 men, sixty-five) women) cruited from hospital staff and volunteers. Their mean systolic blood pressure was 121 (± 15) mmHg, mean diastolic pressure 79 (±9) mmHg. None of the subjects had symptoms of peripheral arterial disease, Raynaud’s phenomenon, or anemia or were receiving any vasoactive and
sex
drugs. In order to look at the influence of disease duration, fasting blood glucose and HbA,, levels, and background retinopathy, the morphologic and hemodynamic parameters in nailfold capillaries within type I diabetics were compared in the following subgroups: Group A: Diabetics with disease duration of more than (n=14) or less than or equal to (n =11 ) fifteen years. Group B: Diabetics with background retinopathy (n=13) or without retinopathy (n=12).
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
397
Group C:
Group
D:
Diabetics with fasting blood glucose level more than (n=10) or less than or equal to (n =15) 8 mmoll-’ . Diabetics with at least two HbAlc values more than (n =11 ) or less than or equal to
(n =14) 7 .0 % . The data of each diabetic
subgroup
were
additionally compared
with those of the
healthy
control group.
Glycosylated hemoglobin was measured by the technique of high-pressure liquid chromatography. The normal range in our laboratory is 3 . 5-5 .5 % . Plasma glucose was measured by a glucose oxydase method. Patients were studied in the morning after having taken their normal breakfast and insulin dose. The finger nailfold capillaries were studied with the help of an incident-light microscope attached to a television monitor.6 The diameters of the nailfold capillaries were measured by means of the 10 x objective before hemodynamic evaluation was begun. The density of the capillary loops represents the number of capillaries per millimeter in the first row of loops parallel to the epidermis. To measure the capillary blood cell velocity (CBV), the television
pictures were videotaped and analyzed during playback by the flying spot technique as previously described.’ Studies were performed following acclimatization during thirty minutes in a room maintained at constant temperature (23°C). CBV was measured at each session five times.’ The first measurement (baseline) was done under normal natural fingertip temperature (V 1 ) . The second reading was performed after the fingertip had been immersed in a warm water bath (40 ° C) for three minutes (V2). The third measurement was done after the observed skin area had been cooled for sixty seconds by blowing rapidly decompressed carbon dioxide of about -15°C over the nailfold (V3). The fourth reading was done after one minute (V4), and the fifth measurement, finally, after two minutes of spontaneous recovery (V5). In cases where the blood flow ceased during local cooling, the duration of the blood flow standstill was measured in seconds. Skin temperature in the investigated area was measured with a thermocouple. Data Analysis Results have been expressed as the mean and the standard deviation. To examine for statistically significant differences between control, study patients, and subgroups, the Mann-Whitney U test (two-tailed) was used. Frequency distribution of nailfold capillary density was determined by contingency tables (Chi square). The Kendall’s rank correlation coefficient was 7
used to determine correlations.
Results Pattern values for the diameters of the arteriolar and venular limbs of the capillary loops (p = 0.01 ) , as well as the capillary width (p = 0.02), showed that these were significantly larger in type I diabetics than in healthy controls (p < 0.05; Table II). The frequency distribution for capillary density showed a significant reduction (p = 0.0001 ) in diabetic patients as compared with healthy control subjects (Table II). Capillary density did not correlate with age, duration of diabetes, HbA,~, or retinopathy. Retinopathy, on the other hand, correlated with the duration of the disease (tau = 0 . 3 8 ; p = 0 . 020) .
Morphology The
mean
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
398 Morphologic Findings
*Number of nailfold
capillary loops
on
TABLE II in Type I Diabetics
Nailfold Capillaries
(Values Are MeanstSD)
per millimeter
Blood Cell Velocity CBV at rest (V 1 ) and after local cooling (V3) were not significantly different from healthy controls or from the values in the different subgroups (Table III). An examination of the mean values of the blood velocities after warm water bath (V2) and in the spontaneous recovery (V4, V5) revealed no conspicuous differences. We found a flow stop with cold exposure in 5/25 patients with type I diabetes (mean duration 9.1 ± 18 . 5 s) and in 1/25 normal controls (12 s). Flow stop in diabetics was not correlated with a lower skin temperature or a history of cold hands.
Capillary
TABLE III Skin
Temperature (°CJ and Capillary Blood Flow Velocity Before (Vi) and After Cold Test (V3) in Type I Diabetics (Values Are Means f SD)
Skin Temperature There was no significant difference in skin temperature between healthy controls and all type I diabetic patients. However, patients with retinopathy tended to have lower skin temperature (p=0.05) than diabetics without retinopathy (Table III). Discussion In our study we found that the nailfold capillaries of type I diabetics differed from those of normal subjects both in an enlargement of the arterial and venular limb and in a significant reduction of the capillary loop density. Retinopathy as major manifestation of the microvascular disease in diabetes mellitus did not correlate with capillary loop density. Our findings of
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
399 abnormalities in diabetics are in agreement with those of Bollinger et a1,9 who fluorescence used videomicroscopy, as well as with those of Fagrell et al,5 Lagrue et al,’° and Tubiana-Rufi et al,&dquo; who found morphologic irregularities by vital nailfold microscopy in such
morphologic patients.
Within type I diabetics we were not able to distinguish between the gross capillary loop morphology of the different subgroups. This may be due to the small number of patients. The comparison of the diabetic subgroups with the healthy control subjects suggests that the enlargement of the morphologic parameters is attributable to the patients with retinopathy, because the mean values of the morphologic parameters of the healthy nondiabetics and the diabetics without retinopathy are nearly equal. Our study has not identified any statistical differences in CBV by nailfold videomicroscopy between diabetic patients and normal control subjects either at rest or after local cooling. In addition, our analyses showed no significant changes of CBV at rest and after local cold exposure in relation to the different fasting blood glucose levels or HbA,~ values. It must be considered, however, that our small group of diabetic patients is rather homogeneous and, except for the patients with retinopathy, nearly free of microangiopathy. Similar velocities at rest have been obtained by Tooke et al’2 in diabetic patients with various disease durations and levels of diabetic control. Some authors suggest that the high skin blood flow observed during poor diabetes control is mainly determined by an increase in arteriovenous shunt flow. 13.14 On the other hand, some authors have found differences in the functional behavior of skin microcirculation when measuring reactive hyperemia following the release of one minute’s digital arterial occlusion or when modifying skin blood flow by venous occlusion.&dquo; These studies were, however, performed with the technique of laser Doppler flowmetry, which measures subpapillary plexus and arteriovenous shunt flow in addition to nutritive capillary flow.’6 Consequently it is not surprising that the results for CBV did not coincide with those from laser Doppler flowmetry, particularly in diabetics, in whom of peripheral blood flow in favor of nonnutritive shunts may be a feature of the partitioning 13.14 disease. Conclusion It may be concluded that the noninvasive technique of vital nailfold capillaroscopy has allowed a differentiation of morphologic parameters between type I diabetics and normal control subjects. In contrast to retinal fluorescein angiography and fluorescein videomicroscopy, the method of nailfold capillaroscopy without dye is simple and noninvasive. However, in spite of these advantages, the concept that the skin microvascular response to a brief period of local cooling may be a useful test for distinguishing complicated from uncomplicated diabetics, as well as from normal controls, could not be confirmed in this study, where the data base may have been too small and rather homogeneous.
Acknowledgments This
study was supported by the Swiss National Science Foundation, grant no. 32.9367.87.
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
400 The authors thank Mrs. B. Dubler for her excellent technical assistance and Mr. S. Christeller, Hoffmann-La Roche, Basel, for statistical advice and for making helpful suggestions.
Paul Gasser, M.D., F.I. C. A. Department of Internal Medicine St. Claraspital Kleinriehenstrasse 30 CH-4016 Basel, Switzerland
References 1. McMillan DE: The microcirculation: Changes in diabetes mellitus. Mayo Clin Proc 63:517-520, 1988. 2. Spiro RG: Glycoproteins and diabetic micro-angiopathy. In: Marble A, White P, Bradley RF, et al, eds. Joslin’s Diabetes Mellitus. Philadelphia: Lee & Febiger, 1971, p 146. 3. Barnes AJ: Blood viscosity in diabetes mellitus. In: Lowe GDO, Barbanel JC, Forbes CD, eds. Clinical Aspects of Blood Viscosity and Red Cell Deformability. New York: Springer Verlag, 1981, p 151. 4. Chitre AP, Velaskar DS: Role of platelets in diabetic microangiopathy—an additional factor. Angiology
5.
5:458-465, 1989. Fagrell B, Hermansson IL, Karlander SG,
11.
12.
et al: Vital
capillary microscopy for assessment of skin viability and microangiopathy in patients with diabetes mellitus. Acta Med Scand 687:25-28, 1984. Bollinger A, Fagrell B: Clinical Capillaroscopy. Toronto : Hogrefe-Huber, 1990, pp 1-30. 7. Boss Ch, Schneuwly P, Mahler F: Evaluation and clinical application of the flying spot method in nailfold capillary TV-microscopy. Int J Microcirc: Clin Exp 6:15-23, 1987. 8. Gasser P: Capillary blood cell velocity in finger nailfold: Characteristics and reproducibility of the local cold response. Microvasc Res 40:29-35, 1990. 9. Bollinger A, Frey J, Jäger K, et al: Patterns of diffusion through skin capillaries in patients with long-term 6.
diabetes. N 10.
13.
14.
15.
16.
Engl J
Med
307:1305-1310, 1982.
Lagrue G, Hamon P, Maurel A, et conjonctivales et périunguéales dans
al: Angioscopies les diabètes noninsulino-dépendants. Valeurs informationelles des anomalies observées. J Mal Vasc 13:101-105, 1988. Tubiana-Rufi N, Priollet P, Lévy-Marchal C, et al: Detection by nailfold capillary microscopy of early morphologic capillary changes in children with insulin dependent diabetes mellitus. Diabète & Métabolism Paris 15:118-122, 1989. Tooke JE, Lins P-E, Ostergren J, et al: Skin microvascular autoregulatory responses in Type I diabetes: The influence of duration and control. Int J Microcirc: Clin Exp 4:249-256, 1985. Boulton AJM, Scarpello JHB, Ward JD: Venous oxygenation in the diabetic neuropathy foot: Evidence of arteriovenous shunting. Diabetologia 22:6-8, 1982. Edmonds ME, Roberts VC, Watkins PJ: Blood flow in the diabetic neuropathic foot. Diabetologia 22:9-15, 1982. Tooke JE, Ostergren J, Lins P-E, et al: Skin microvascular blood flow control in long duration diabetics with and without complications. Diabetes Res 5:189-192, 1987. Tooke JE, Ostergren J, Fagrell B: Synchronous assessment of human skin microcirculation by laser Doppler flowmetry and dynamic capillaroscopy. Int J Microcirc : Clin Exp 2:277-284, 1983.
Downloaded from ang.sagepub.com at The University of Iowa Libraries on June 24, 2015
![[Ocular vasospasm. 2: Diagnosis of vasospastic syndrome using nailfold capillary videomicroscopy].](/assets/img/hold.png)
