Developing evidence-based clinical guidelines for military use: case study of smoking cessation guidelines.

OBJECTIVE: To develop evidence-based clinical guidelines on smoking cessation, for use throughout the British military. METHOD: A ten-member, multiprofessional smoking cessation working group met five times between October 2000 and July 2001 to develop targeted smoking cessation guidelines for use by military health professionals in the clinical setting. The guidelines were based on the best available scientific evidence at that time, mainly systematic review of controlled trials, and individual randomised trials. RESULTS: The agreed military guidelines on smoking cessation were promulgated in July 2001. Three tiers of support were defined. Military health professionals have a key role as nonsmoking models and advocates, and should be trained to use 'brief intervention' at every clinical encounter with a military smoker. 'Intermediate support' (defined as a specialist service delivered by military health professionals who have undergone specific training and continuation training) is to be available at local level. The most heavily addicted military smokers will require referral to civilian smoking cessation clinics. Effective technologies for use at any one of the three levels of care are: nicotine skin patches, nicotine gum, nicotine lozenges and bupropion. CONCLUSIONS: These are the first ever clinical guidelines for military use which meet accepted modern quality criteria. Informal monitoring of the uptake of these guidelines between July 2001 to December 2001 suggests that they have been well received by military health professionals. An audit of their impact on smoking patterns within the UK Armed Forces will commence in 2002. The guidelines will be updated 5-yearly, or sooner.

Does military service damage females? An analysis of medical discharge data in the British armed forces.

There is anecdotal and some scientific evidence that females in military service experience an excess of work-related injuries, compared with males. To investigate this more fully, we analysed data collected routinely by the Defence Analytical Services Agency on medical discharges in male and female personnel in the British armed forces. We found that for all disease and injury categories of medical discharge there is a statistically significant excess in females; this disparity is particularly marked for discharges on account of injury [relative risk (RR) = 1.65, 95% confidence interval (95% CI) = 1.30-2.10] and musculoskeletal disease (RR = 3.34, 95% CI = 2.75-4.06). Royal Navy females are eight times more likely (RR = 7.92, 95% CI = 3.03-20.66) and Army females seven times more likely (RR = 6.53, 95% CI = 2.60-16.42) than Royal Air Force females to be medically discharged on account of injury. Over the period 1993-1996, there was a statistically significant increase in the rate of medical discharge for both musculoskeletal disease and injury in female personnel in the British armed forces. During the period 1996-2000, a marked gender differential was maintained, but the rate of increase in females reached a plateau. We concur with previous investigators that mixed-sex training imposes particular ergonomic stresses on females and that it is a major risk factor for overuse injury. We discuss other possible explanations for the marked gender differential in medical discharge rates in the military. Some changes to training programmes are now being introduced to correct this health inequality, but further interventions are needed. Modifications to training programmes must be audited systematically and candidate interventions tested through randomized controlled trials.

The malaria threat.

The British experience of the past two decades indicates that fixed-schedule chemoprophylaxis is difficult because of the variety of epidemiological situations and increasing incidence of drug resistance. Of the 110 cases of malaria contracted in Kenya between 1982 and 1996, 74% were due to Plasmodium falciparum. Of the 45 malaria infections contracted in Belize, 84% were due to Plasmodium vivax. In 1985 the fixed drug combination of chloroquine base 300 mg weekly plus proguanil 200 mg daily was adopted as standard chemoprophylaxis for use in all parts of the world where chemoresistant Plasmodium falciparum had been observed. Mefloquine was recommended as first-line prophylaxis in Papua New Guinea in 1986 and in East Africa in 1993. Doxycycline hyclate was prescribed in September 1999 when a Gurkha company was deployed on peacekeeping duties to East Timor, but its effectiveness has not yet been evaluated. Chemoprophylaxis must be combined with non-drug antimalaria technologies, especially insecticide-treated bed nets.

Wave-front measurement errors from restricted concentric subdomains.

In interferometry and optical testing, system wave-front measurements that are analyzed on a restricted subdomain of the full pupil can include predictable systematic errors. In nearly all cases, the measured rms wave-front error and the magnitudes of the individual aberration polynomial coefficients underestimate the wave-front error magnitudes present in the full-pupil domain. We present an analytic method to determine the relationships between the coefficients of aberration polynomials defined on the full-pupil domain and those defined on a restricted concentric subdomain. In this way, systematic wave-front measurement errors introduced by subregion selection are investigated. Using vector and matrix representations for the wave-front aberration coefficients, we generalize the method to the study of arbitrary input wave fronts and subdomain sizes. While wave-front measurements on a restricted subdomain are insufficient for predicting the wave front of the full-pupil domain, studying the relationship between known full-pupil wave fronts and subdomain wave fronts allows us to set subdomain size limits for arbitrary measurement fidelity.

Heparin management test versus activated coagulation time during cardiovascular surgery: correlation with anti-Xa activity.

OBJECTIVE: To compare the abilities of the heparin management test (HMT) and the activated coagulation time (ACT) to provide a measurement of heparin effect in patients undergoing cardiac or peripheral vascular surgery. These measurements of heparin effect were also compared with measurements of heparin concentrations tested by anti-Xa activity. A secondary objective was to compare the performance of the noncitrated HMT with that of the citrated HMT. DESIGN: A prospective study. SETTING: A single-center study conducted in a university hospital. PARTICIPANTS: After human investigation committee approval and informed consent were obtained, adult patients undergoing cardiac or peripheral vascular surgery were included in this study. INTERVENTIONS: In both surgical groups, blood was sampled for ACT, HMT, and anti-Xa activity. Each HMT was performed on both noncitrated and citrated samples. MEASUREMENTS AND MAIN RESULTS: As an indicator of heparin effect, the HMT had a strong correlation with the ACT (r = 0.899; p < 0.01). In addition, the HMT had a significantly stronger correlation with anti-Xa activity than the ACT (p < 0.01). The correlation obtained from the noncitrated samples was identical with that obtained from the citrated samples (r = 0.819; p < 0.001 for both groups). CONCLUSION: The ability of the HMT and the ACT to measure heparin effect was similar. The HMT performed better than the ACT when using anti-Xa activity as a measure of heparin concentration. Noncitrated HMT results were similar to citrated HMT results, thus supporting the use of fresh whole blood for testing purposes.

Point-of-care and standard laboratory coagulation testing during cardiovascular surgery: balancing reliability and timeliness.

OBJECTIVE: The use of point-of-care technology has increased faster than efforts to validate its effectiveness compared to standard laboratory testing modalities. To address this issue with a current point-of-care coagulation system (HEMOCHRON Jr, International Technidyne Corporation (ITC), Edison, NJ), we designed a study to test the hypothesis that data obtained from point-of-care coagulation equipment correlates with data obtained from standard laboratory coagulation equipment. One of the potential advantages gained using point-of-care testing is the ability to obtain more rapid results. To address this issue, turnaround time, defined as the elapsed time (in minutes) from when the sample was acquired from the patient until the investigators knew the results, was also determined. METHODS: Following Human Investigation Committee approval and informed consent, a prospective study was conducted to compare results obtained from point-of-care coagulation equipment with those results obtained from standard laboratory coagulation equipment. The study was performed in three groups of patients undergoing cardiovascular surgery, each requiring different levels of anticoagulation. RESULTS: Of the 83 patients who met the inclusion criteria, the correlation (combining data from groups 1-3) between results obtained from point-of-care and standard laboratory prothrombin time was r = 0.867, p < 0.001. The correlation (group 3) between point-of-care and standard laboratory international normalized ratio was r = 0.943, p < 0.001. The correlation (combining data from groups 1 & 2) between point-of-care and standard laboratory activated partial thromboplastin time was r = 0.825, p < 0.001. Median turnaround time for the standard laboratory was 90 minutes, with a mean turnaround time of 74 to 78 minutes, depending upon the group. In contrast, the median turnaround time for point-of-care testing was two minutes and 14 seconds. CONCLUSIONS: The results from this study population reveal that data obtained from point-of-care prothrombin time, international normalized ratio and activated partial thromboplastin time results correlate with results obtained from standard laboratory coagulation testing. The value of obtaining reliable results in a timely fashion offers a potential advantage for point-of-care testing in dinical situations, such as in the operating room, where saving time may translate into financial savings.

Comparison of streptokinase, urokinase, and recombinant tissue plasminogen activator in an in vitro model of venous thrombolysis.

PURPOSE: Presumed differences in the thrombolytic activity and fibrinolytic specificity of the three commonly used thrombolytic agents, streptokinase, urokinase, and recombinant tissue plasminogen activator (rt-PA), are based on clinical study results, where variability renders meaningful comparisons difficult. An in vitro model of catheter-directed venous thrombolysis was used to compare the three agents. METHODS: Retracted iodine 125-radiolabeled clots that simulate those observed in the venous system were infused with thrombolytic agents at doses analogous to those used clinically. Perfusion with heparinized, whole human blood was undertaken for 60 minutes, measuring the efficacy of thrombolysis through serial quantification of radio tracer released into the circuit. Fibrinolytic specificity was determined by following decrements in perfusate fibrinogen concentration. RESULTS: Streptokinase was the agent associated with the slowest rate of clot lysis (p = 0.01 vs urokinase and rt-PA). Urokinase was associated with an intermediate rate of lysis but appeared to be the agent with the greatest degree of fibrinolytic specificity (p = 0.02 vs streptokinase, p = 0.05 vs rt-PA). Although rt-PA was associated with improved efficacy early in the perfusions, the differences between rt-PA and urokinase dissipated after 30 minutes. CONCLUSIONS: These laboratory observations suggest that urokinase may be the most appropriate agent for catheter-directed venous thrombolysis, offering an advantageous compromise between fibrinolytic specificity and thrombolytic speed.

Angiotensin II type 1 receptor: role in renal growth and gene expression during normal development.

To determine whether angiotensin II (ANG II) modulates renal growth and renin and angiotensin type 1 (AT1) gene expression via AT1 during development, weanling rats were given ANG II antagonist losartan (DuP 753) for 3 wk. Body weight (g), kidney weight (g), and kidney weight-to-body weight ratio were lower in losartan-treated rats (162 +/- 7, 1.6 +/- 0.06, and 9.5 +/- 0.1 x 10(-3)) than in control rats (184 +/- 5, 1.8 +/- 0.07, and 10.1 +/- 0.1 x 10(-3); P < 0.05). Renal DNA content (mg/kidney) was lower in losartan-treated (2.4 +/- 0.17) than in control rats (3.3 +/- 0.31; P < 0.05), whereas protein-to-DNA and RNA-to-DNA ratios were similar in losartan-treated and control rats. Renin mRNA levels were sevenfold higher in losartan-treated than in control rats, as determined by quantitative standardized dot blot analysis. In addition, blockade of AT1 with losartan induced recruitment of renin-synthesizing and renin-containing cells in the renal vasculature, as determined by immunocytochemistry and in situ hybridization. To establish whether AT1 blockade has a direct effect on renin gene expression, freshly isolated renin-producing cells were exposed in vitro to losartan (10(-6) M) or culture media (control). Losartan induced a twofold increase in steady-state renin mRNA levels above control (P < 0.05). Intrarenal AT1 mRNA levels were not altered by losartan given either in vivo or in vitro to freshly dispersed cells. To define whether immature renin-secreting cells are responsive to ANG II, renin release was determined by reverse hemolytic plaque assay.(ABSTRACT TRUNCATED AT 250 WORDS)

Neonatal ureteral obstruction stimulates recruitment of renin-secreting renal cortical cells.

Unilateral ureteral obstruction (UUO) in the neonate increases ipsilateral renal renin gene expression, an effect which is mediated by renal nerves. To determine whether neonatal UUO alters the number of renal cortical cells secreting renin and whether this change is modulated by renal nerve activity, newborn Sprague-Dawley rats were subjected to left UUO, right uninephrectomy, or sham operation and studied four weeks thereafter. To evaluate the importance of renal nerves in this response, an additional group of animals underwent chemical sympathectomy with guanethidine. Ureteral obstruction was associated with marked reduction in renal mass in the obstructed kidney and contralateral compensatory hypertrophy, changes which were not altered by sympathectomy. Renin messenger RNA and renal renin content were elevated in the obstructed kidney. The number of cells secreting renin, measured by the reverse hemolytic plaque assay, was markedly increased in the obstructed kidney (45 +/- 18 plaques/slide vs. 11 +/- 1 plaques/slide in sham animals), but not in the opposite kidney or following uninephrectomy. This effect was not significantly altered by sympathectomy. There was no change in the amount of renin secreted per cell or in the secretory response to Ca++. These results show that UUO results in recruitment of cells not previously secreting renin by a mechanism independent of renal nerve activity. This recruitment occurs without alteration of the quantity of renin secreted per cell or in the normal regulatory effect of Ca++ on renin secretion. An increase in the number of renin-secreting cells may contribute to the activation of the renin-angiotensin system, and thus to the vasoconstriction observed following ureteral obstruction.

Neurologic events following carotid endarterectomy: prediction of outcome.

A total of 1572 carotid endarterectomies were performed at one institution between 1975 and 1987. One hundred five patients had early (< 3 weeks) neurologic events following carotid endarterectomy. Sixty-five patients had cerebral vascular accidents (CVAs) (4.1%), 14 patients had reversible ischemic neurologic deficits (0.9%), and 26 patients had transient ischemic attacks (1.7%). Eight patients died from CVAs (0.5%). The mean follow-up was 31 months (range 1 to 137 months) with a 5-year cumulative survival of 77%. The median time of occurrence of neurologic events was 4 hours. Ages, cerebral protection, patches, carotid occlusion time (mean 29 minutes), gender, and status of the contralateral carotid arteries were not predictors of outcome. Death from neurologic events increased significantly in patients who had preoperative CVAs compared with patients with preoperative transient neurologic deficits (p < 0.05). The time of occurrence of CVA after carotid endarterectomy affected outcome, and an early CVA (< 4 hours) was associated with a higher mortality at 30 days and at 4 months as a consequence of the initial CVA (p = 0.11). Patients who had a neurologic event more than 4 hours after surgery had a significantly better resolution of their symptoms (66%) compared with patients who had an early neurologic event (35%, p < 0.05). The long-term follow-up of the surviving patients demonstrated an improvement in neurologic function in 75% of the CVA group (36/48) and 92% (76/83) of all patients who had neurologic events in long-term follow-up.

Effects of angiotensin converting enzyme inhibition, sodium depletion, calcium, isoproterenol, and angiotensin II on renin secretion by individual renocortical cells.

Angiotensin-converting enzyme inhibition with enalapril increases the number of glomeruli with juxtaglomerular cells and the number of cells in the afferent arteriole that express the renin gene and contain renin. However, renin release from these newly recruited renin-containing cells has not been demonstrated. Sodium depletion also has been shown to increase renal renin messenger RNA levels. The aim of these studies was to determine whether increases in renin secretion are a result of altered numbers of cells synthesizing/releasing renin or a change in the amount of renin release per cell, or both. Adult Wistar-Kyoto rats were treated with enalapril or sodium depleted and single cell renin secretion of enzymatically dispersed renal cortical cells was examined by reverse hemolytic plaque assay. Enalapril treatment increased the number of renin secreting cells by approximately 10-fold (P < 0.05). The newly recruited renin-secreting cells were not responsive to changes in extracellular calcium concentration or the presence of isoproterenol. At physiological (2.5 mM) extracellular calcium concentration, the amount of renin secreted per cell was approximately 2-fold greater (P < 0.05) when cells from enalapril-treated rats were compared to controls and sodium depletion increased both the number of renin-secreting cells and the amount of renin secreted by approximately 35% (P < 0.05). Angiotensin II (AII) inhibited the number of cells secreting renin in cortical cells prepared from enalapril-treated and control rats. In conclusion, angiotensin converting enzyme inhibition increased renin secretion predominantly by recruitment of additional renin-secreting cells and, to a lesser extent, by augmentation of the amount of renin released per cell. In contrast, sodium depletion increased renin secretion equally by both mechanisms. Newly recruited renin-secreting cells were not regulated by the extracellular calcium concentration or beta-adrenergic stimulation. Angiotensin II inhibited renin secretion directly by decreasing the number of individual cells releasing renin through a process which was independent of the extracellular calcium concentration.

Colocalization and release of angiotensin and renin in renal cortical cells.

Angiotensin is generated within the kidney, but the precise loci for the formation of angiotensin I (ANG I) and angiotensin II (ANG II) have not been demonstrated. We performed electron microscopy immunocytochemistry in kidney sections of 10-day-old (newborn) and adult Wistar-Kyoto (WKY) rats using specific antibodies to renin, ANG I, ANG II, and angiotensinogen (AO). Renin, ANG I, ANG II, and AO were present in juxtaglomerular (JG) cells. Renin was largely confined to cytoplasmic granules; ANG I and ANG II were colocalized to these granules but also were present in the cytoplasm; AO was distributed throughout the cytoplasm. AO also was present in a renal cortical distribution in proximal tubular cells. Northern blot analysis demonstrated AO mRNA in total kidney and liver but not in renal microvessels. Using the reverse hemolytic plaque assay, we demonstrated release of ANG I and renin from individual renocortical cells of adult WKY rats. Under control conditions, the number of releasing cells was 11 +/- 1 for ANG I and 10 +/- 1 for renin. Addition of rat renin inhibitor (RI) (1 x 10(-5) M), which inhibited renin activity in the medium from 37 to 9 pg ANG I.ml-1.h-1, did not alter ANG I plaque number. Addition of rat AO increased ANG I plaque number to 17 +/- 2 (P less than 0.05). Incubation with both RI and AO prevented the increase in ANG I plaque number obtained with AO alone. Enalapril treatment (7 days; n = 5) increased the number of plaque-forming cells to 22 +/- 2 for ANG I (P less than 0.0005) and to 39 +/- 7 for renin (P less than 0.001). The results suggest an intracellular location for AO and angiotensin and release of renin and ANG I by renal cortical cells and suggest that released angiotensin is produced intracellularly and that secretion of ANG I is augmented by converting enzyme inhibition.

Fate of the exposed saphenous vein graft.

Patients with functioning vein grafts may present with wound problems resulting in exposure of the graft. We reviewed the courses of 16 patients presenting with this problem to determine the causative factors and the most appropriate management of this therapeutic dilemma. Diabetes mellitus (11 of 16, 68.7%) and wound infection (12 of 16, 75%) were frequent pre-existing conditions associated with exposed venous grafts. All patients with exposed vein grafts were initially treated conservatively with regular application of moist sterile dressings, followed by split-thickness skin graft coverage of the wounds when clean. The wounds healed in 7 patients, whereas 9 patients developed complications of hemorrhage (7 patients) and graft thrombosis (2 patients). The outcome of therapy was highly dependent on the type of organism originally cultured from the wounds. The incidence of vein graft disruption was lowest when the wounds were sterile (25%) or when gram-positive bacteria grew (25%). Gram-negative infection uniformly resulted in disruption of the exposed venous graft. When a new graft was placed, the secondary graft became reinfected in all patients with gram-negative primary graft infection. There were no instances of secondary graft reinfection when gram-negative bacteria were not present. These data suggest that the outcome of patients presenting with exposed vein grafts is highly dependent on the bacterial flora of the process. Vein graft disruption is frequent in patients with gram-negative infection, suggesting that these patients should be treated with distant graft ligation and extra-anatomic bypass. By contrast, patients without gram-negative infection may be successfully managed with local wound care.

Factors determining survival after ruptured aortic aneurysm: the hospital, the surgeon, and the patient.

The 10-year experience of a single community was reviewed and a multivariate analysis was performed to determine the relative importance of clinical and environmental factors in mortality after ruptured abdominal aortic aneurysm resection. Ruptured aneurysms were repaired in 243 patients in six area hospitals (one university, five community) by 25 surgeons (16 vascular, 9 general). Overall, 30-day mortality was 55% (133/243). Although the mortality by hospital ranged from 44% to 68%, these differences were not statistically significant. However, significant variations occurred in the mortality rates of individual surgeons, ranging from 44% to 73%. The mortality rate for the vascular surgeons was less than that of the general surgeons, 51% versus 69% (p less than 0.05). Clinical factors were evaluated, and the most significant parameters were systolic blood pressure, presence of chronic obstructive lung disease, and history of chronic renal insufficiency. These results support the implication that the degree of specialization of the surgeon and the preexisting health of the patient are the most important determinants of survival after ruptured abdominal aortic aneurysm. The size and sophistication of the hospital appear to be less influential factors.

Identification of individual renocortical cells that secrete renin.

Successful application of the reverse hemolytic plaque assay was developed to identify individual renocortical cells that secrete renin directly. The plaque assay was validated by a number of established criteria. Using this technique, we demonstrate an increase in renin secretion with beta-adrenergic stimulation and an inhibition of renin secretion with extracellular calcium in groups of renin-secreting cells. Transmission electron microscopy of the cell in the center of a hemolytic plaque demonstrated a modified vascular smooth muscle cell with densely packed secretory granules. Electron microscopy immunocytochemistry demonstrated the presence of renin in the secretory granules, confirming the identity of the cell as a renal juxtaglomerular cell. The technology developed here has allowed the precise identification and study of the individual renin-secreting juxtaglomerular cell.

Differential effects of a gram-negative and a gram-positive infection on autogenous and prosthetic grafts.

A canine model was developed to study the differential response of a gram-negative and a gram-positive bacterial infection on autogenous and prosthetic grafts. After replacing segments of the femoral arteries of 15 dogs with autogenous vein in one groin and polytetrafluoroethylene in the contralateral groin, 10(8) colony-forming units of nonmucin-producing Staphylococcus epidermidis (five dogs), Pseudomonas aeruginosa (five dogs), or sterile saline solution (five dogs) were directly inoculated onto the grafts. The grafts were examined 7 to 10 days after implantation. None of the control dogs exhibited inflammatory signs, and no grafts or anastomoses disrupted. S. epidermidis was unrecoverable from either graft material in any of the animals, although histologic evaluation confirmed neutrophils and bacteria in four of five animals in the vein and polytetrafluoroethylene groups. No dog inoculated with S. epidermidis had graft or anastomotic disruption. By contrast, P. aeruginosa was recovered from both types of grafts in all inoculated animals. Neutrophils, bacteria, and microabscesses were observed in all of these animals. In addition, three of five polytetrafluoroethylene grafts and all five vein grafts disrupted either at the anastomoses or in the body of the vein graft. Therefore S. epidermidis is a less virulent organism that may persist in graft walls despite negative cultures, whereas P. aeruginosa is a highly virulent organism that can disrupt native artery, vein grafts, and anastomoses. The graft material appears to be less important than the bacteria in determining the outcome of infection.