Julian A.J.H. Critchley

A Meta-Analysis of Studies Using Bias and Precision Statistics to Compare Cardiac Output Measurement Techniques

Introduction. Bias and precision statistics have succeeded regression analysis when measurement techniques are compared. However, when applied to cardiac output measurements, inconsistencies occur in reporting the results of this form of analysis. Methods. A MEDLINE search was performed, dating from 1986. Studies comparing techniques of cardiac output measurement using bias and precision statistics were surveyed. An error-gram was constructed from the percentage errors in the test and reference methods and was used to determine acceptable limits of agreement between methods. Results. Twenty-five articles were found. Presentation of statistical data varied greatly. Four different statistical parameters were used to describe the agreement between measurements. The overall limits of agreement in studies evaluating bioimpedance (n = 23) was ±37% (15–82%) and in those evaluating Doppler ultrasound (n = 11) ±65% (25–225%). Objective criteria used to assess outcome were given in only 44% of the articles. These were (i) limits of agreement approaching ±15–20%, (ii) limits of agreement of less than 1 L/min, and (iii) more than 75% of bias measurements within ±20% of the mean. Graphically, we showed that limits of agreement of up to ±30% were acceptable. Conclusions. When using bias and precision statistics, cardiac output, bias, limits of agreement, and percentage error should be presented. Using current reference methods, acceptance of a new technique should rely on limits of agreement of up to ±30%.

Toxicity of Complementary Therapies: An Eastern Perspective

Chinese traditional medicine is used extensively in Chinese populations, and other Asian countries employ similar therapies. Herbal treatments have a major role in these systems, and although most have a well‐established safety record, occasional adverse effects are seen. Problems arise when toxic herbs are used in excessive doses, with improper preparation, or when they are substituted erroneously. There may also be adulteration with Western drugs or heavy metals, and interactions between herbs and Western drugs may also occur. It is always prudent to obtain a complete history of the use of herbal medications during any clinical assessment, particularly in Asian patients.

Drug-induced disorders of glucose metabolism mechanisms and management

SummaryGlucose homeostasis is maintained by a balance between the release and action of insulin, and the counterregulatory responses mediated principally by glucagon, catecholamines, growth hormone and cortisol. Hence, the effects of a drug on glucose metabolism may be mediated by any of these agents singly or in combination. Host factors, such as inherent glucoregulatory mechanisms, concurrent diseases, organ function and concomitant medications also increase the risk of drug-induced disturbances of glucose homeostasis in susceptible individuals. By far the most important agents causing hypoglycaemia are insulin and the sulphonylureas. Alcohol (ethanol), over-zealous glycaemic control, hypoglycaemic unawareness, defective counterregulation especially in insulin-dependent diabetes mellitus (IDDM), and renal and liver impairment are all important predisposing factors. Although antihyperglycaemic agents such as metformin and α-glucosidase inhibitors do not cause hypoglycaemia alone, they may enhance the hypoglycaemic effects of potent hypoglycaemic agents such as insulin and sulphonylureas. On the other hand, the potential hypoglycaemic effects of ACE inhibitors, α-blockers, lipid-lowering agents and recombinant human insulin-like growth factor demonstrated in experimental settings, are of potential therapeutic interest.Iatrogenic hypoglycaemia and intensive insulin treatment are associated with hypoglycaemic unawareness which may be obviated by meticulous avoidance of hypoglycaemia. Effective patient education remains an important preventive measure. Oral glucose is used to treat mild hypoglycaemic episodes while more severe episodes are treated by intravenous glucose or glucagon. Nasal glucagon and theophylline are other experimental measures to improve recovery from hypoglycaemia. In refractory hypoglycaemia due to hyperinsulinaemia such as during sulphonylurea overdosage or quinine treatment, the long-acting somatostatin, octreotide, may suppress insulin release and restore euglycaemia.Diuretics, β-blockers, sympathomimetics, corticosteroids and sex hormones are commonly prescribed drugs which may have adverse effects on carbohydrate metabolism especially in patients with diabetes mellitus or those who are at risk of developing glucose intolerance. Pentamidine was frequently associated with dysglycaemia due to its pancreatic β-cell cytotoxic effects but is now used less often to treat Pneumocystis carinii pneumonia in immunosuppressed patients.Despite the large number of anecdotal reports of drug-induced disturbances of glucose metabolism, many of the so-called adverse drug reactions were either idiosyncratic or coincidental. Nevertheless, they emphasise the complex nature of glucose homeostasis and its potential interactions with drugs, host factors and disease states. An understanding of these relationships may allow more critical interpretation of these clinical observations, better prediction of drug-induced adverse effects on carbohydrate metabolism and the implementation of more rational therapy. Hence, the hypoglycaemic effects of a drug may be turned to a therapeutic advantage in patients with glucose intolerance. Similarly, the hyperglycaemic effect of a drug may help to treat refractory hypoglycaemia.

Usage and adverse effects of Chinese herbal medicines

The great majority of Chinese herbal preparations are safe, and in the past, some useful Western drugs have been derived from these herbs. Nearly all serious poisonings are due to the few preparations containing aconitine, podophyllin or anti cholinergics or else proprietary preparations containing dangerous Western drugs or heavy metals. Both medical professionals and the general public should be alerted to the potential toxicity of herbal remedies. There should be frequent monitoring of Chinese herbal medicines or their derivatives, such as some Chinese proprietary medicines, for undeclared Western drugs and heavy metals. Mothers should be discouraged from treating their children with herbal or proprietary medicines. There should be con tinuing efforts to collect safety information on these widely used products.

Gastric emptying following brain injury: Effects of choice of sedation and intracranial pressure

ObjectiveTo compare the effects of opioid and nonopioid sedation on gastric emptying.DesignProspective, randomized trial.SettingUniversity teaching hospital ICU.Patients21 brain injured patients requiring sedation, mechanical ventilation and intracranial pressure (ICP) monitoring for >24h.InterventionsPatients were randomized to receive infusions of either morphine plus midazolam (M), or propofol (P). Gastric emptying was assessed by the paracetamol absorption technique and by residual volumes following a 200 ml test feed.measuerments and resultsPre-sedation Glasgow Coma Score, mean ICP and the presence of bowel sounds were noted. Plasma concentrations of paracetamol were measured over 3 h following a 1 g gastric dose. There were no differencese in median peak paracetamol concentration (M, 18.5 versus P, 20.8 mg/l), median time to peak concentration (M, 20 versus P, 25 min), median area under the concentration-time curve (AUC), or in the median residual volumes at 1 h (M, 14 versus P, 10.5 ml) and 2 h (M, 5 versus P, 3 ml). In patients with ICP>20 mmHg, paracetamol concentrations were lower (p<0.05), and AUC after 30 min was lower (165 mg·min/l versus 411 mg·min/l,p=0.023). Mean ICP was correlated with AUC (Kendall rankp=0.027). Gastric emptying did not correlate with initial Glasgow Coma Score or presence of bowel sounds.ConclusionsGastric emptying is not improved in patients with brain injury by avoiding morphine (1–8 mg/h) in the sedative regimen. Intracranial hypertension is associated with reduced gastric emptying.

Antihypertensive and anti-albuminuric effects of losartan potassium and felodipine in Chinese elderly hypertensive patients with or without non-insulin-dependent diabetes mellitus

After a 4-week placebo baseline period, 29 Chinese elderly hypertensive patients were randomized, double-blind, to 12 weeks of treatment with either losartan potassium (n = 19), an angiotensin II antagonist at the AT1 receptor, or felodipine (n = 10), a calcium channel blocking agent. Of these 29 patients 12 had coexisting non-insulin-dependent diabetes mellitus. At week 12, the mean reductions (95% confidence intervals) in mean arterial pressure were similar in both groups: losartan -18 (range -22 to -14) mm Hg; felodipine -19 (range -25 to -11) mm Hg. In the whole group, the 24-hour urinary albumin excretion was reduced by 27% with losartan as compared with no change in the felodipine-treated group (p = 0.03; analysis of variance). In the diabetic group, losartan treatment reduced the urinary albumin excretion by 24% as compared with 11% in the felodipine-treated group. In the non-diabetic patients, the urinary albumin excretion fell by 29% in the losartan-treated group, but increased by 14% in the felodipine-treated group (p < 0.001; repeated-measures analysis of variance). Plasma sodium increased to a similar extent in both groups. The fasting plasma triglyceride level declined by 25% (p < 0.001 within group) with losartan, but was not significantly reduced in the felodipine-treated group. For comparable reductions in blood pressure, a greater reduction in albuminuria was seen with losartan than with felodipine treatment in Chinese hypertensive patients with or without non-insulin-dependent diabetes mellitus. Long-term studies are required to examine whether these antiproteinuric effects of losartan can be translated to renoprotection.

Metabolic and Hemodynamic Effects of Metformin and Glibenclamide in Normotensive NIDDM Patients

Objective— To compare the metabolic and hemodynamic effects of metformin and glibenclamide in normotensive NIDDM patients. Research Design and Methods— After a 2-wk run-in period on dietary treatment alone, 12 Chinese normotensive patients with uncomplicated NIDDM were randomized to receive either metformin, or glibenclamide for 4 wk before being crossed-over to the alternative treatment for an additional 4 wk. Metabolic and hemodynamic index, including cardiac output estimation by impedance cardiography, were measured at baseline and at the end of each treatment period. Results— Body mass index was reduced more with metformin than with glibenclamide, although glycemic control was similar with both drugs. Plasma total cholesterol concentration fell more with metformin (mean difference −0.65 mM, 95% confidence interval −0.96 to −0.32) than glibenclamide (mean difference −0.20 mM, 95% confidence interval −0.54–0.12) (P < 0.05). Compared with baseline values, erect diastolic blood pressure was reduced more by metformin (12.9% [95% confidence interval −21.5 to −4.4%]) than glibenclamide (−6.8% [95% confidence interval −14.9 to 1.2%]) (P < 0.001). The relative changes in the systemic vascular resistance index also differed between the two treatments (glibenclamide, 6.2 [−4.3 to 16.6%]; metformin, −1.2 [95% confidence interval −8.8−6.4%]) (P < 0.05)]. Conclusions— In normotensive NIDDM patients, treatment with metformin was associated with greater reductions in body weight, plasma total cholesterol concentration, and erect diastolic blood pressure, whereas the systemic vascular resistance index increased after treatment with glibenclamide. These findings merit long-term investigation.

The estimation of paracetamol and its major metabolites in both plasma and urine by a single high-performance liquid chromatography assay

Many analytical methods exist for the assay of paracetamol in biological fluids, including colorimetry with chemical derivatization, direct spectrophotometry, chromatographic methods and immunoassays. Their development has been largely driven by the needs of clinical toxicology requiring the rapid, reliable and highly specific estimation of paracetamol in plasma samples to determine the need for antidote therapy. However, for in vivo metabolism studies, a specific assay method which can provide measurements of paracetamol and its metabolites in both plasma and urine is desired. A reversed-phase HPLC method with UV detection at 254 nm was developed to fulfil these requirements. The assay involves minimum sample preparation with a relatively short run time. The solvent system involves a simple isocratic elution with a composition of 0.1 M potassium dihydrogen orthophosphate-acetic acid-propan-2-ol, (100:0.1:0.75, v/v/v). The reproducibility of the assay was high with an inter-assay RSD of 0.2-1.7% for urinary paracetamol concentrations of 5-500 micrograms ml-1 and 0.1-3.3% for plasma concentrations between 5 and 25 micrograms ml-1. A similarly high degree of precision was found for the glucuronide, sulphate, cysteine and mercapturate metabolites of paracetamol. The same assay can be used to analyse both plasma and urine samples and thus was employed for studies on the metabolism of paracetamol in healthy subjects and in patients with various diseases.

Visceral Fat and Cardiovascular Risk Factors in Chinese NIDDM Patients

OBJECTIVE The interrelations between obesity, glucose intolerance, hypertension, dyslipidemia, and insulin resistance are well recognized. These relationships are of particular interest in Hong Kongs Chinese population, in whom increasing affluence has coincided with a marked increase in the prevalence of NIDDM. We designed a pilot study to examine the relationships between visceral fat and cardiovascular risk factors in Chinese NIDDM patients. RESEARCH DESIGN AND METHODS We studied 21 Chinese NIDDM patients whose visceral fat was quantified by magnetic resonance imaging. Cardiovascular risk factors including plasma lipids and 24-h ambulatory blood pressure (BP) were measured. In addition, insulin resistance was determined by a short insulin tolerance test (SITT). RESULTS Increased visceral adiposity was significantly correlated with plasma triglycerides (r = 0.63, P = 0.004), the total cholesterol/HDL cholesterol ratio (r = 0.61, P = 0.008), the urinary albumin/creatinine ratio (r = 0.49, P = 0.04), and decreased insulin sensitivity as measured by the SITT (r = 0.47, P = 0.03). When the data were analyzed by tertiles, increasing visceral fat area was associated with higher plasma triglycerides, lower HDL cholesterol, and a smaller plasma glucose decrement during the SITT. In addition, the diurnal rhythm in BP and heart rate tended to be best preserved in those with the least visceral obesity. CONCLUSIONS This pilot study demonstrates that visceral fat accumulation is associated with dyslipidemia, hypertension, insulin resistance, and albuminuria in Chinese patients with NIDDM.

Angiotensinogen T235 and ACE Insertion/Deletion Polymorphisms Associated With Albuminuria in Chinese Type 2 Diabetic Patients

OBJECTIVE Genetic polymorphisms of the renin-angiotensin system (RAS) have been implicated in the pathogenesis of diabetic proteinuria. Ethnic differences in the frequencies of these genotypes have also been reported. To date, most of these studies have been performed in white and Japanese populations. In this study, we examined the associations between albuminuria and RAS genetic polymorphisms in Chinese patients with type 2 diabetes. RESEARCH DESIGN AND METHODS In a case-control study, the ACE insertion/deletion (I/D) gene, the angiotensinogen (AGT) gene (M235T), and the angiotensin II (AII) type 1 receptor gene (AT1 A1166C) were examined in 110 Chinese type 2 diabetic patients. Increased urinary albumin excretion (UAE) was defined as ≥ 30 mg/day on at least two occasions during a 6-week study period. RESULTS Compared with whites, there were high frequencies of the AGT TT genotype in Chinese control subjects (120/183 = 70%) and type 2 diabetic patients (74/110 = 67%). The frequencies of the MM genotype were 5 and 3%, respectively, and those of the ACE DD genotype were 13 and 10%, respectively. Although 9% of subjects carried the C allele, the AT1 CC genotype was not found in either group. Chinese type 2 diabetic patients with increased albuminuria (n = 56) had higher systolic blood pressure (160 ± 26 mmHg vs 145 ± 27 mmHg, P < 0.001) than the normoalbuminuric patients (n = 54). Both the AGT TT genotype (78.6% [44/56] vs. 55.6% [30/54], odds ratio [OR]: 3.0 [1.3–6.8]) and the T allele (88% [99/112] vs. 77% [83/108], OR: 2.5 [1.3–5.4]) were associated with an increased risk of albuminuria. Patients with the AGT TT genotype (n = 74) had higher 24-h UAE than those with the MT or MM genotypes (n = 36) (median: 37.8 mg/day vs. 17.8 mg/day, P < 0.01). This association remained significant in patients with normotension (56 mg/day [n = 19] for patients with the TT genotype vs. 22 mg/day [n = 14] for those with the MT/MM genotype, P = 0.03). The D allele carriers (DD or DI, n = 61) had higher serum ACE activities (75.5 ± 29 U/l vs. 60.5 ± 36.3 U/l, P < 0.01) than the noncarriers (II genotype). The median 24-h UAE also tended to be higher in the D allele carriers (38.9 mg/day vs. 21.4 mg/day, P = 0.07). The lowest UAE was observed in patients with the MM/MT/II genotype (16.3 mg/day [n = 18]) and the highest, in patients with the TT/DD/DI genotype (52.3 mg/day [n = 43]). No association was found between the TT genotype or D allele and hypertension. CONCLUSIONS The high frequencies of the TT genotype and T allele in Chinese populations may contribute to the high prevalence of albuminuria in patients with type 2 diabetes. The possibility of synergism between the AGT TT genotype and the ACE D allele should also be explored.