番茄社区

Patients with chronic kidney disease (CKD), including those with diabetes, who were not on dialysis were more likely to die when blood pressure was too low, even if only on the diastolic measure, a veterans cohort showed.

The relationship between systolic blood pressure and mortality followed a U-shaped curve that bottomed out at 130 to 159 mm Hg, while the J-shaped curve for diastolic pressure hit its optimal point at 70 to 89 mm Hg, , of the VA Medical Center in Memphis, Tenn., and colleagues found.

Considered together, coming in a little high on systolic blood pressure appeared to be less risky than going too low with diastolic pressure, the researchers reported in the Aug. 20 issue of the .

The likelihood of death for patients with an optimal systolic pressure of 130 to 139 mm Hg rose the lower the diastolic pressure went below 70 mm Hg. Adjusted hazard ratios were (all P<0.05):

• 1.09 at 60 to 69 mm Hg
• 1.14 for 50 to 59 mm Hg
• 1.37 for 40 to 49 mm Hg
• 1.91 for anything below 40 mm Hg

By contrast, for patients in the optimal diastolic range of 70 to 89 mm Hg, mortality risk changed relatively little as systolic pressure rose. Hazard ratios were 1.04 to 1.19 as blood pressure went up to 180 to 189 mm Hg and not significant beyond that.

"It may not be advantageous to achieve ideal systolic blood pressure at the expense of lower-than-ideal diastolic blood pressure in adults with chronic kidney disease," the researchers argued, adding "we suggest caution in lowering blood pressure to less than what has been demonstrated as beneficial in randomized, controlled trials."

The recommend a treatment target of less than 140/90 mm Hg in CKD without proteinuria or less than 130/80 mm Hg if there is micro- or macroalbuminuria.

"These targets emphasize the lowering of systolic pressure, without considering the potential clinical effects of low diastolic blood pressure," Kovesdy's group noted. "Guidelines for lowering systolic blood pressure in patients with albuminuria will probably result in even lower diastolic blood pressure with treatment."

However, suggest caution in trying to reach the systolic goal of less than 140 mm Hg if diastolic pressure goes down to a "potentially dangerous level less than 65 mm Hg."

The study in CKD patients "reinforces an important note of caution for practicing physicians that in order to provide more protection to more patients, we must not concomitantly and unintentionally induce harm to a very large population of individuals at risk," , of the University of Miami and chief of nephrology hypertension at the Miami VA Medical Center, commented in an email to 番茄社区. Raij was not involved in the study.

The study took advantage of the large database of the VA healthcare system for a closer look at every combination of systolic and diastolic blood pressure -- 96 different categories in all -- in 10 mm Hg increments in relation to all-cause mortality over a median 5.8 years of follow-up.

Among the 651,749 veterans (mostly male; only 2.7% female) with CKD not dependent on dialysis, 43% had diabetes.

The highest mortality risk was at those extremes of blood pressure: about six-fold for less than 80/40 mm Hg and two-fold for more than 210/120 mm Hg.

Subgroup analyses didn't suggest different results for normal versus elevated urinary microalbumin-creatinine ratios.

The researchers suggested compromised blood flow to vital organs as a reason for the risk associated with too-low blood pressure, "especially low diastolic blood pressure compromising coronary perfusion," but acknowledged that confounding by stiff arteries or the high burden of comorbidities could have played a role.

Because the observational results couldn't determine causality, prospective randomized trial evidence is needed to clarify the best treatment goals in CKD and to see if active treatment improves survival in people with low diastolic pressure, they added.

"Specifically, we cannot conclude that the mortality risk associated with various blood pressures in our study is equal to the risk imparted by the same blood pressures when they occur as a result of antihypertensive interventions in clinical practice," the group cautioned.

Other limitations were the almost-exclusively male population, the potential for residual confounding, the relatively low proportion of patients with measurements of albuminuria, and use of clinic-measured rather than ambulatory or research-quality measurements of blood pressure.

Comment