Mortality retrospective studies

Mortality associated with glucocorticoid has been retrospectively studied in 556 patients with chronic obstructive pulmonary disease admitted to a rehabilitation center (353). Median survival was 38 months and 280 patients died during follow-up. On multivariate analysis, oral glucocorticoid use at a prednisone equivalent of 10 mg/day without inhaled glucocorticoid was associated with an increased risk of death (RR = 2.34 95% Cl = 1.24, 4.44), and 15 mg/day increased the risk further (RR = 4.03 95% Cl = 1.99, 8.15). The risk of death was not increased in those using 5 mg/day or when patients used any oral dose in combination with inhaled glucocorticoids. 

In retrospective studies, neither the degree of hyperlac-tatemia nor accumulation of metformin had prognostic significance, but mortality was linked to the underlying disease (79,80). 

Contrast-induced nephropathy has been defined as an increase in serum creatinine of at least 25% or an absolute increase in serum creatinine of at least 0.5 mg/dL within 48 to 72 hours of iodinated contrast administration and is associated with significant morbidity and mortality (75). Important risk factors include diabetes mellitus, chronic renal insufficiency, administration of large volumes of high osmolar contrast agents, and intravascular volume depletion. Numerous pharmacologic preventive measures have been studied, but consistent benefits have not been demonstrated. In a recent large retrospective study, preprocedural statin therapy was independently associated with a lower risk of contrast nephropathy and nephropathy requiring dialysis (76). 

From the available literature it is impossible to have a proper evaluation of the aluminum levels in the water used for dialysis, the serum levels in patients and their outcome, due to insufficient data and retrospective studies. Water aluminum values can vary substantially over a short period, and some water companies may either be reluctant to release their data, or did not test aluminum on a frequent basis. There have been only two outbreaks of acute encephalopathy one in Portugal, and one on Curasao with similar symptomatology and a mortality above 30%. In both studies aluminum levels in the water used for dialysis was measured after the outbreak and in both studies levels above 650 pg/L were found. In several cases levels above 1000 pg/L, and retrospectively calculated up to 8400 pg/L were reported, without similar clinical symptoms. Because acute aluminum encephalopathy has been extremely rare and with regard to the fact that patients seem to tolerate extremely high serum levels (above 1000 pg/L) before acute aluminum encephalopathy develops, it seems likely that in outbreaks in hemodialysis, levels of the water used for dialysis need to be at least above 1000 to 1500 pg/L before acute aluminum encephalopathy can develop. 

Cheng WN, Kong J. 1992. A retrospective mortality cohort study of chrysotile asbestos products workers in Tianjin 1972-1987. Environ Res 59(l) 271-278. 

In a retrospective study of 64 patients, mean age 71 years, with acute renal insufficiency associated with an ACE inhibitor, over 85% presented with overt dehydration due to diuretics or gastrointestinal fluid loss (69). Bilateral renal artery stenosis or stenosis in a solitary kidney was documented in 20% of cases. In seven patients dialysis was required, but none became dialysis dependent. After resolution of acute renal insufficiency, the plasma creatinine concentration returned to baseline and renal function was not significantly worsened. Two-year mortality was the highest in a subgroup of patients with pre-existing chronic renal insufficiency. 

In a retrospective study of 153 patients who underwent bone-marrow transplantation for chronic myelogenous leukemia, pretransplant interferon alfa treatment for more than 12 months was associated with a significant increase in transplant-related mortality during the first 2 years when compared with patients who received pretransplant interferon alfa for less than 12 months (28 of 46 patients versus nine of 38) (384). This adverse outcome was also more frequent in patients who discontinued treatment less than 3 months before transplantation. 

Potassium chloride is irritating to the gastrointestinal tract, even to the extent of causing perforation (4). In a retrospective study at surgical clinics in Stockholm County there were 22 cases of small-bowel ulceration in which a connection with slow-release potassium chloride tablets was probable (5). Most of the ulcers had caused stenosis of 1-2 cm of gut, and in four cases there was also perforation of the bowel wall. Five patients had perforation without signs of stenosis. Mortality was 27%. The pathology of the ulcers was similar to that described after use of enteric-coated potassium chloride tablets. The frequency of potassium-induced ulceration is low (about 3 cases per 100 000 patient-years of slow-release tablet use), but this complication can be serious. 

However, accumulating evidence supports the use of norepinephrine in patients with septic shock with a retrospective study demonstrating reduced mortality with norepinephrine over other vasopressors [106]. Furthermore, animal data demonstrates that reversal of septic hypotension with norepinephrine leads to increases in renal blood flow [107]. There are no studies that compare the renal outcomes between catecholamine therapy and vasopressin. 

A proportionate mortality analysis (PMR) revealed an increase of leukemia deaths in gasoline station attendants and mechanics. In addition, there were four deaths due to brain cancer (Schwartz 1987). No information is provided on length or employment or exposure. The PMR does not reflect the mortality of the population because it is so easily influenced by over- and under-representation of deaths. The automobile mechanics and gasoline station attendants who died of leukemia each had different job titles. Therefore, they may not have had similar exposures. Also, several types of leukemia were found acute myelogenous leukemia (N=2), chronic myelogenous leukemia (N=2), erythroleukemia (N=1), chronic lymphocytic leukemia (N=2), acute leukemia unspecified (N=1), and chronic leukemia undifferentiated (N=1). It is very difficult to draw any definitive conclusions from this type of analysis because several different types of leukemia were reported, and quantification of exposure is not possible in a retrospective study of this type. 

A retrospective study of poisoning admissions to the six major referral hospitals in Zimbabwe over a 10-year period (1980-1989) revealed 6018 cases. The main cause (23%) was from acute poisoning by traditional medicines. Total mortality was 15%, with traditional medicines being the second highest canse of fatalities 80). 

Yusho and Yu-Cheng Exposures. The retrospective study Yusho victims summarized in Section 3.2.8.2.1 found no statistically significant (p<0.05) increased mortality from cancer of the stomach or esophagus (Kuratsune et al. 1987). The retrospective study of Yu-Cheng victims summarized... 

Mortality from cancer of the brain and nervous system was not statistically significantly different than expected in Kimbrough et al. (1999a) and Gustavsson and Hogstedt (1997) retrospective studies of capacitor manufacturing workers summarized in Section 3.2.8.2.I. 

Death Deaths associated with commonly prescribed antipsychotic drugs [SEDA-32, 89 SEDA-33, 90] have been assessed in a 5-year retrospective study using US Veterans National Healthcare data [54 ]. The exposed subjects, who were predominantly male, aged 65 and older and with a diagnosis of dementia, were compared with randomly selected controls. Those who were exposed to haloperidol (n = 2217), olanzapine ( = 3384), quetiapine (n = 4277), or risperidone ( = 8249) had more co-morbidities than the controls. During the first 30 days, there was a significant increase in mortality in those who took doses of haloperidol over 1 mg/day (HR=3.2 95% Cl = 2.2, 4.5), olanzapine over 2.5 mg/day (HR = 1.5 95% CI=1.1, 2.0), or risperidone over 1 mg/day... 

---