Surgery and trauma care providers' perception of the impact of dual-practice employment on quality of care provided in an Andean country.

BACKGROUND: Dual-practice, simultaneous employment by healthcare workers in the public and private sectors is pervasive worldwide. Although an estimated 30 per cent of the global burden of disease is surgical, the implications of dual practice on surgical care are not well understood. METHODS: Anonymous in-depth individual interviews on trauma quality improvement practices were conducted with healthcare providers who participate in the care of the injured at ten large hospitals in Peru's capital city, Lima. A grounded theory approach to qualitative data analysis was employed to identify salient themes. RESULTS: Fifty interviews were conducted. A group of themes that emerged related to the perceived negative and positive impacts of dual practice on the quality of surgical care. Participants asserted that the majority of physicians in Lima working in the public sector also worked in the private sector. Dual practice has negative impacts on physicians' time, quality of care in the public sector, and surgical education. Dual practice positively affects patient care by allowing physicians to acquire management and quality improvement skills, and providing incentives for research and academic productivity. In addition, dual practice provides opportunities for clinical innovations and raises the economic status of the physician. CONCLUSION: Surgeons in Peru report that dual practice influences patient care negatively by creating time and human resource conflicts. Participants assert that these conflicts widen the gap in quality of care between rich and poor. This practice warrants redirection through national-level regulation of physician schedules and reorganization of public investment in health via physician remuneration.

Locally produced ready-to-use supplementary food is an effective treatment of moderate acute malnutrition in an operational setting.

BACKGROUND: Typical treatment of moderate acute malnutrition, simple wasting, in sub-Saharan Africa consists of dietary counselling and/or general or targeted distribution of corn/soy-blended flour (CSB). A randomised clinical effectiveness trial in 2007 showed CSB to be less effective than ready-to-use supplementary food (RUSF). AIM: To determine the operational effectiveness of treating moderate acute malnutrition with RUSF. METHODS: Children aged 6-59 months were recruited in rural southern Malawi. Each child received 65 kcal/kg/d of locally produced soy/peanut RUSF, a product that provided about 1 RDA of each micronutrient. Anthropometric measurements were taken every 2 weeks and additional rations of RUSF were distributed at this time if the child remained wasted. Study participation lasted up to 8 weeks. RESULTS: Of the 2417 children enrolled, 80% recovered, 4% defaulted, 0.4% died, 12% remained moderately wasted and 3% developed severe acute malnutrition. Weight, length and MUAC gain were 2.6 g/kg/d, 0.2 mm/d and 0.1 mm/d respectively. Cost per child treated was $5.39. CONCLUSIONS: This intervention proved to be robust, maintaining high recovery rates and low default rates when instituted without the additional supervision and beneficiary incentives of a research setting.

Randomized, multicenter comparison of oral granisetron and oral ondansetron for emetogenic chemotherapy.

STUDY OBJECTIVES: To compare the antiemetic effectiveness and safety of oral granisetron plus dexamethasone with those of oral ondansetron plus dexamethasone administered before emetogenic chemotherapy. DESIGN: Randomized, prospective, multicenter, open-label study. SETTINGS: University-teaching hospital and veterans health care system. PATIENTS: Sixty-one chemotherapy-naïve patients scheduled to receive emetogenic antineoplastic agents. INTERVENTION: A single-dose oral granisetron 1 mg and dexamethasone 12 mg or single-dose oral ondansetron 16 mg and dexamethasone 12 mg was administered before chemotherapy. MEASUREMENTS AND RESULTS: Twenty-four hours after administration patients were contacted to assess nausea, emesis, and adverse events. There were no statistical differences in frequency of nausea or emesis between groups. Seventy-six percent and 82% of patients receiving ondansetron and granisetron, respectively, experienced no emesis 24 hours after chemotherapy. Complete protection from nausea occurred in 58% and 46% of patients receiving the drugs, respectively. Adverse events were similar between groups. CONCLUSION: Oral granisetron 1 mg and ondansetron 16 mg plus dexamethasone are safe and effective in preventing nausea and vomiting related to emetogenic chemotherapy.

Stat proteins play a role in tumor necrosis factor alpha gene expression.

Trauma produces dysfunction in immunity, which appears to be partially related to alterations in the cytokine response. Signal transducer and activator of transcription proteins (STATs) mediate activation of several cytokine genes. However, the effect of STAT proteins on tumor necrosis factor-alpha (TNFalpha) activation is not fully defined. We identified binding sites for STAT 3 and STAT 5/6 within the promoter region of TNFalpha and hypothesize that alterations in these sites would affect TNFalpha expression. The TNFalpha promoter was inserted into the luciferase reporter vector, and binding sites for STAT 3, STAT 5/6, and activator protein-1 (AP-1) were mutated using site-directed mutagenesis. Murine macrophages were transfected with the resultant plasmids, then incubated with and without lipopolysaccharide (LPS) or IFNalpha. Gene expression was measured by dual luciferase assay. Mutation of the STAT 3 binding site was associated with decreased LPS-inducible activity. Mutation of the AP-1 and STAT 5/6 consensus binding sites alone had no effect on TNFalpha expression. However, combined mutation of both STAT 5/6 and AP-1 was associated with increased LPS-inducible activity. Mutations of the STAT binding sites in the promoter region of TNFalpha affect TNFalpha gene expression. These results suggest a regulatory role for STATs in TNF gene transcription.

Inhibition of leukocyte-mediated tissue destruction by synthetic fibronectin peptide (Trp-9-Tyr).

Burns are surrounded by an inflammatory zone of stasis that can progress to ischemia and extension of burn size. Synthetic fibronectin peptides have reduced tissue destruction in several models of inflammation. In this study, we postulate that administration of the peptide Trp-9-Tyr will alter the progression of tissue destruction following thermal injury. Baseline cutaneous blood flow was measured on New Zealand White rabbits with a laser doppler blood-flow meter. While the rabbits were under general anesthesia, 6 full-thickness burns were produced on the rabbits' backs. Blood flow in the zones of stasis was followed daily, and the number of zones that progressed to necrosis was determined at 72 hours. There were 3 experimental groups. Ten control animals received saline. Ten were treated with Trp-9-Tyr for 24 hours postburn. Ten received Trp-9-Tyr for 48 hours. Animals treated with Trp-9-Tyr had higher blood flow and less necrosis in the zones of stasis than did control animals, which was evident at 24 hours but more significant at 48 hours.

Enhanced GFR response to oral versus intravenous arginine administration in normal adults.

Both oral protein ingestion and intravenous amino acid infusions have been shown to increase glomerular filtration rate (GFR) and renal plasma flow (RPF) in normal subjects. Although the mechanism of this effect is not known, the renal responses to these loads have been associated with increases in peripheral glucagon concentrations. Conflicting data exist concerning the role of glucagon in the hyperfiltration response after an oral protein meal or administration of an intravenous amino acid mixture. Using a single amino acid as the stimulus for hyperfiltration, we compared the renal responses in six normal subjects to 30 gm oral arginine-HCl, intravenous arginine-HCl, and intravenous glucagon infused at the rate of 10 ng/kg/min. GFR, RPF, and glucagon concentration, as well as levels of plasma amino acids and selected gastrointestinal hormones, were measured for six 30-minute clearance periods after each load. Significant rises in mean peak GFR were noted after both oral arginine (104 +/- 5 ml/min x 1.73 m2 to 145 +/- 9 ml/min x 1.73 m2, p less than 0.02) and intravenous arginine (118 +/- 10 ml/min x 1.73 m2 to 134 +/- 11 ml/min x 1.73 m2, p = 0.02) administration. Mean peak RPF rose significantly after oral arginine (510 +/- 26 ml/min x 1.73 m2 to 710 +/- 32 ml/min x 1.73 m2, p less than 0.01) but not after intravenous arginine (616 +/- 60 ml/min x 1.73 m2 to 687 +/- 64 ml/min x 1.73 m2, p = 0.18). Intravenous glucagon infusion also increased both mean peak GFR (99 +/- 9 ml/min x 1.73 m2 to 149 +/- 10 ml/min x 1.73 m2, p less than 0.01) and RPF (514 +/- 48 ml/min x 1.73 m2 to 771 +/- 38 ml/min x 1.73 m2, p less than 0.01) significantly. We found the mean peak percent rise in GFR (43% +/- 13%) and RPF (42% +/- 12%) after oral arginine to be notably greater than that after intravenous arginine (14% +/- 5% and 13% +/- 9%, respectively). However, the mean peak percent rise in glucagon concentration after oral arginine was significantly lower than that after intravenous arginine (62% +/- 25% versus 479% +/- 176%, respectively, p = 0.04). Infusion of glucagon increased GFR (54% +/- 13%) and RPF (55% +/- 12%) to a degree similar to that seen after oral arginine, but again with a significantly higher mean peak percent rise in peripheral glucagon concentrations when compared with the rise after oral arginine (798% +/- 348% vs 62% +/- 25%, p less than 0.05).(ABSTRACT TRUNCATED AT 400 WORDS)

Effect of dietary protein restriction on functional renal reserve in diabetic nephropathy.

PURPOSE, PATIENTS, AND METHODS: Functional renal reserve in patients with insulin-dependent diabetes mellitus, as determined by the glomerular filtration rate (GFR) response test, is a measure of the capacity of the kidney to increase glomerular filtration in response to the stimulus of a protein meal or amino acid infusion. This 12-month study evaluated the changes in functional renal reserve in eight patients with insulin-dependent diabetes mellitus with nephropathy (micro-albuminuria [greater than or equal to 30 micrograms/minute]) who chronically decreased their dietary protein intake to a mean of 0.6 g/kg/day (Group 1) compared with a group of similar patients (n = 7) who maintained their unusual dietary protein intake (1.0 g/kg/day, Group 2). Patients were evaluated and measurements taken at 3-, 6-, and 12-month intervals. Absolute and percent increases in GFR were calculated from three averaged 1-hour measurements after an 80-g protein test meal. RESULTS: Although the initial absolute mean rise (14 +/- 12 versus 18 +/- 13 mL/minute/1.73 m2) in GFR and maximal percent rise (16% +/- 16% versus 32% +/- 27%) after the meal did not differ significantly between the two groups, at 12 months, values in the lower protein group increased (27.8 +/- 9.5 mL/minute/1.73 m2 and 54.7% +/- 48.8%), whereas those in the normal protein intake group declined significantly (3.7 +/- 3.6 mL/min-ute/1.73 m2 and 6.5% +/- 6.5%) (p less than 0.05). Both urine urea and microalbuminuria decreased significantly (p less than 0.05) in the low protein group. Unstimulated GFR at the end of 12 months was significantly less (p less than 0.05) in Group 2 (47 +/- 2 mL/minute/1.73 m2) than in Group 1 (71 +/- 21 mL/minute/1.73 m2). The rate of decline in GFR was significantly greater (p less than 0.05) in the normal protein intake group than in the low protein intake group (0.68 +/- 0.4 versus 0.28 +/- 0.15 mL/minute/1.73 m2/month). CONCLUSIONS: This study indicates that sustained dietary protein restriction can help to preserve renal function, decrease albuminuria, and lower the baseline GFR while maintaining functional renal reserve in patients with insulin-dependent diabetes mellitus.

Effect of indomethacin on the glomerular filtration rate after a protein meal in humans.

The renal response to a protein meal has been characterized by increases in glomerular filtration rate (GFR) and renal plasma flow and decrease in renal vascular resistance. Several hormonal mediators of this response have been proposed, including renal prostaglandins (PGs). We studied ten normal subjects before and after ingestion of indomethacin. All subjects had three 30-minute baseline creatinine and iothalamate clearances measured before and three one-hour clearances measured after an 80-g protein meal. The night before the second test, the subjects took 25 mg indomethacin and 150 mg one hour before the test meal. Urine PG excretion decreased significantly during the second test, from 0.60 +/- 0.23 to 0.30 +/- 0.14 ng/min (P less than 0.01). Initial iothalamate clearance increased from 110 +/- 10 to a mean of 122 +/- 15 mL/min/1.73 m2 (average increase of 12 mL/min/1.73 m2) during the second hour after the test meal. After ingestion of indomethacin, the GFR remained unchanged from a baseline of 101 +/- 9 to 101 +/- 7 mL/min/1.73 m2 (average change of -1 mL/min/1.73 m2). Because the time of peak increase after the meal varied from subject to subject, the maximal increase in GFR after the meal was calculated and found to be significantly less during the second test, 13 +/- 3 v 22 +/- 4 mL/min/1.73 m2 (P less than 0.05). Because PGs can stimulate glucagon secretion and because glucagon has been suggested to mediate the protein-stimulate GFR response, we measured plasma glucagon during both tests; these levels were not different.(ABSTRACT TRUNCATED AT 250 WORDS)

Somatostatin limits rise in glomerular filtration rate after a protein meal.

To evaluate the glomerular filtration rate (GFR) response to a protein meal in patients with diabetes and to study the role of glucagon and growth hormone, we studied inulin clearance for three 30-minute periods before and 3 hours after an 80 g protein meal in seven healthy volunteers and 10 patients with diabetes. Patients with diabetes were chosen because their renal response to such a meal has been reported to be abnormal. All had an increase in GFR and plasma glucagon levels after the protein meal. The peak rise in GFR occurred from 1 to 2 1/2 hours after the meal (mean +/- SEM, delta 26 +/- 5 mL/min/m2, controls; delta 22 +/- 7 mL/min/m2, patients with diabetes), with the mean time to normal rise in GFR occurring at 2 hours after the meal. Similarly, plasma glucagon values peaked at different times in individual patients (delta 769 +/- 532 pg/mL, controls; delta 267 +/- 69 pg/mL, patients with diabetes), with the mean plasma glucagon rise occurring 1 hours after the meal. Premeal growth hormone levels tended to be higher in the patients with diabetes (7.6 +/- 1.4 vs 2.1 +/- 0.4 ng/mL), and did not change after the meal. To allow study of the contribution of the increased plasma glucagon to the rise in GFR, eight of these patients (five with diabetes) volunteered to undergo a second GFR response test with a simultaneous infusion of somatostatin. The glucagon response was significantly lowered at all time periods during the infusion (P less than 0.05); no significant change in growth hormone occurred. Without somatostatin in these eight patients, peak increase in postmeal GFR average 20.6 +/- 1.5 mL/min/m2; with the somatostatin, peak increase in GFR was 6.0 +/- 1.8 mL/min/m2 (P less than 0.01). Neither metabolic control nor degree of albuminuria was significantly different at the time of the two studies. Thus, as has been shown in animals, somatostatin infusion limits the rise in GFR after a protein meal in humans.

Effect of varied protein intake on the nephropathy of the diabetic mouse (C57BL/s J db/db).

Protein intake has been suggested to influence progression of renal disease by affecting intraglomerular pressures and flows. The renal disease of the diabetic mouse (C57BL/Ks/db/db) has been proposed as a suitable model of human diabetic nephropathy. Ten diabetic mice and ten non-diabetic controls were placed on 1 of 3 protein intakes, 4%, 27% and 50%, and serial functional measurements were made at 2 to 3 week intervals until week 20. All diabetic animals showed similar degrees of hyperglycemia. The creatinine clearances were generally higher in the diabetic mice than the controls, except the 4% protein intake diabetic group, until week 20 when the 27% db/db mice showed a significant decline (p less than 0.05) compared to the control mice. Albumin excretion was significantly higher in the 27% and 50% protein intake db/db mice than the controls. Again the 4% group showed albuminuria not different from the control animals. Histologic studies at 20 weeks showed minimal abnormalities in the normal and 4% protein intake diabetic group. The 27% and 50% intake diabetic mice showed a progressive increase in severity of mesangial matrix expansion with segmental sclerosis. Electron microscopy confirmed these findings. Immunofluorescence microscopy showed a marked increase in mesangial immunoglobulin G and M. With similar degree of hyperglycemia, higher protein intake was associated with more severe histologic changes, greater albuminuria and early decline in GFR. Thus protein intake can markedly affect the progression of renal disease in the diabetic mouse.

Short-term protein loading in diabetics with a ten-year duration of disease.

The concept of renal functional reserve has recently been introduced. To test this ability of the kidneys to increase the glomerular filtration rate (GFR) above the baseline level, the GFR response to short-term protein load was measured. Recent studies have provided conflicting data concerning the GFR response to a protein load in insulin-dependent diabetics who are known to have increased baseline GFRs. Thus, we studied nine insulin-dependent diabetics with a disease of at least a ten-year duration (none were hypertensive or proteinuric) and compared their data with those of five nondiabetic controls with normal renal function. All the diabetics, except one, showed a significant increase in GFR (mean +/- SEM, 60 +/- 9 to 74 +/- 14 mL/min/sq m); the controls also had increased GFRs (mean +/- SEM, 53 +/- 6 to 69 +/- 6 mL/min/sq m). The one patient who demonstrated no rise in the GFR had the lowest GFR measured, 33 mL/min/sq m. To explore the mechanism of this response, we measured the plasma levels of putative mediators glucagon and human growth hormone. Although glucagon showed the expected rise after the protein meal, the variability was so large that no statistically significant relationship could be identified. Human growth hormone remained constant and low in the controls and showed more variability and was higher in the diabetics; again, no relationship to the GFR could be demonstrated. Thus, our data demonstrated a normal response to a short-term protein load by a group of well-defined diabetic subjects who would be at risk to show subtle renal abnormalities.

Acute response of urinary N-acetyl-beta-D-glucosaminidase to mannitol infusion in the dog.

Urinary activity of N-acetyl-beta-D-glucosaminidase (NAG) has been used as an indicator of subtle renal injury in a variety of conditions. Such enzyme activity has been shown to be increased in human and other animals with diabetes mellitus. The mechanism of this increase in urinary NAG activity is not known. To determine if the osmotic diuretic effect of the glycosuria could stimulate urinary NAG activity, mannitol was infused into the left renal artery of six dogs to cause a unilateral osmotic diuresis and compared to the right side. During three control periods of 20 minutes, each urinary NAG excretion (expressed in units as the ratio of NAG activity to urinary creatinine, NAG/Cr) was equal from both left and right kidneys, 5.0 +/- 1.5 vs 6.0 +/- 3.6 units, respectively. During the 11 mannitol infusion periods urine volume and sodium excretion rose significantly from the left kidney, .50 +/- 2 to 1.5 +/- .3 ml/min and 21 + 5 to 99 +/- 16 u Eq/min, respectively. However urinary NAG/Cr did not change, 5.0 +/- 1.5 to 5.1 +/- 1.0 units. In six control dogs not infused with mannitol, urinary NAG/Cr tended to rise with time from control to experimental collection periods, 4.7 +/- 2.0 to 8.1 +/- 3.0 respectively; however these are not significantly different. In all dogs urine volume and sodium excretion tended to rise throughout the course of the study due to hydration with normal saline; thus it is possible that the tendency for urinary NAG activity to rise may have been due to the increase in sodium excretion. However, these studies demonstrate that the osmotic diuresis induced by mannitol produced no significant change in urinary NAG activity. Thus it may be that the hyperglycemia itself, and not the glycosuria, produces the increase in urinary NAG activity seen in the diabetic.

Differential response of urinary N-acetyl-beta-D-glucosaminidase to two osmotic diuretics in the dog.

Urinary N-acetyl-beta-D-glucosaminidase (NAG), a lysosomal hydrolase located the proximal tubule of the kidney, has been used as a marker for subtle renal injury. In humans and other animals with diabetes mellitus, urinary NAG activity has been shown to increase within 12 hours of the onset of hyperglycemia and glycosuria. Whether the rise in urinary NAG activity is in response to the hyperglycemia or to the osmotic diuresis associated with glycosuria is not known, nor has the time course of the rise in enzyme activity been determined. A study was designed using four groups of dogs to examine these possibilities: group 1 (n = 5), control dogs; group 2 (n = 5), mannitol-infused dogs; group 3 (n = 5), low-glucose dogs; and group 4 (n = 5), high-glucose dogs. In groups 2 and 4, mannitol and glucose, respectively, were infused at a rate to double urine flow from the left ureter without altering the contralateral urine volume. In group 3, sufficient glucose was infused to elevate left renal vein glucose level without producing glycosuria. In the control dogs infused with normal saline solution at a constant rate throughout the control and study periods, no differences were found in urinary NAG excretion when data from individual clearance periods were compared for the right and left kidneys. In the low-glucose dogs, urinary NAG/creatinine ratios were significantly increased (P less than 0.01) when the left and right kidneys were compared for the duration of the infusion.(ABSTRACT TRUNCATED AT 250 WORDS)

Concentrations of prostaglandins in plasma, seminal vesicles, and ovaries of aging C57BL/6NNia mice.

Concentrations of prostaglandin E1 (PGE1), and prostaglandin E2 (PGE2) (combined in the same radioimmunoassay) and prostaglandin F2 alpha (PGF2 alpha) were analyzed in circulating plasma and seminal vesicles of 3- and 26 to 27-month-old males and in circulating plasma and ovaries of 3-, 6-, 14 to 18- and 26 to 30-month-old female C57BL/6NNia mice. The amount of PGE declined in the plasma (P less than 0.05) and seminal vesicles (P less than 0.02) of aged male mice, whereas PGF2 alpha concentrations remained unchanged. There were no statistical differences in plasma or ovarian concentrations of PGE or PGF2 alpha when comparing the various age groups of female mice. It does not appear as if age-related changes in prostaglandins play a significant role in reproductive senescence.

Response of urinary N-acetyl-beta-D-glucosaminidase to rapid decreases in blood glucose.

Urinary N-acetyl-beta-D-glucosaminidase (NAG) has been shown to be a sensitive indicator of blood sugar control. Twelve insulin-dependent diabetics whose blood glucoses were being controlled with the artificial pancreas had concurrent urinary NAG activity measured. Blood glucose dropped markedly from 198 +/- 22 mg/dl (x +/- SEM) to 121 +/- 7 mg/dl during the 25 h on the artificial pancreas. For the entire group UNAG: UCr dropped from 14.9 +/- 4.4 to 7.25 +/- 1.68 units. In order to determine if larger decreases in blood glucoses over the course of the study resulted in larger decreases in UNAG: UCr, an arbitrary division at 180 mg/dl was made. Six patients with blood glucoses at or above this level at the start of the study showed a drop in both blood glucoses and UNAG: UCr (261 +/- 24 to 134 +/- 12 mg/dl and 21.1 +/- 6.1 to 7.29 +/- 1.53 U, respectively). Even though the other six patients had blood glucoses below the renal threshold, both blood glucoses and UNAG: UCr declined (136 +/- 4 to 110 +/- 4 mg/dl; 8.89 +/- 2.4 to 6.2 +/- 1.64 U, respectively). Thus not only may urinary NAG activity reflect long-term control and renal complications of diabetes, but this enzyme is also responsive to acute changes in blood glucose.

Urinary N-acetyl-beta-D-glucosaminidase in streptozotocin-induced diabetic rats.

Excretion of urinary N-acetyl-beta-D-glucosaminidase has been found to be elevated in diabetic humans and rats. This urinary glycosidase may reflect blood sugar control over time, since it has been significantly and positively correlated with hemoglobin A1 in children with insulin-dependent diabetes. Other studies have suggested that urinary NAG may predict diabetic nephropathy. In order to more carefully define the relationship between urinary NAG excretion and blood and urine sugars, hemoglobin A1, and microalbuminuria, 48 rats were made diabetic by the use of streptozotocin. All rats were uninephrectomized at 3 weeks. Of these, 23 were treated with daily insulin injections, 25 were untreated, and both groups were compared to 13 control, nondiabetic rats. Urine volume, glucose, albumin, and blood sugar were all significantly (P less than 0.05) elevated in the untreated rats compared to the treated and control groups. Urinary NAG:UCr was significantly (P less than 0.01) elevated in the untreated group with lower but still elevated levels (P less than 0.05) in the treated rats. To further define the time course of the increase in UNAG:UCr 12 rats were followed serially at 12-hr intervals for 92 hr after streptozotocin. Urinary NAG increased significantly (P less than 0.05) at 12 hr after streptozotocin injection and reached a plateau at 36 hr while hemoglobin A1 did not rise until 2 weeks after onset of hyperglycemia. Urinary NAG increases more rapidly than hemoglobin A1 after onset of hyperglycemia and glycosuria.(ABSTRACT TRUNCATED AT 250 WORDS)

Urinary excretion of N-acetyl-beta-D-glucosaminidase in children with type I diabetes mellitus.

N-acetyl-beta-D-glucosaminidase (NAG), a lysosomal enzyme, has been shown to be increased in the urine of patients with various glomerulonephritides, tubulointerestitial diseases, renal allograft rejection, toxic renal injury, and diabetes mellitus. Although it has been suggested that urinary NAG may reflect blood glucose control, no studies have correlated this with other measures of metabolic control. Thirty-four children from a diabetic summer camp were found to have urinary NAG to creatinine ratios significantly above those of normal controls of similar age (5.22 +/- 1.19 versus 1.51 +/- 0.17 U). Urinary NAG was found to positively correlate with an arbitrary control index (r = 0.82; P less than 0.05) and in seven patients with hemoglobin A1c (r = 0.70; P less than 0.001). In a closely followed group of 40 clinic patients, urinary NAG to creatinine ratio was again found to be significantly increased over normal controls (7.55 +/- 0.70 versus 1.51 +/- 0.17 U; P less than 0.05). Again, urinary NAG was positively correlated with HbA1c (r = 0.62; P less than 0.001) and urinary albumin to creatinine ratio (r = 0.47; P less than 0.01). In neither group was there a correlation with UNAG:UCr and duration of disease. Thus, these data suggest that urinary NAG to creatinine ratio appears to be a reflection of blood sugar control.