Improved long-term cardiovascular outcomes after intensive versus standard screening of diabetic complications: an observational study

Diabetes mellitus is associated with reduced life expectancy owing to severe multi-organ complications. On average, a 50-year-old man without a history of cardiovascular disease (CVD) but with diabetes dies 6 year earlier than a subject without diabetes, with the majority of premature causes of death being attributable to CVD [1]. Despite the global decline in mortality and incidence of CVD over the last decades, seen both in subjects with and without diabetes, the latter still experience a 2–3 times higher risk as compared to non-diabetic matched controls [1‐3]. This higher risk, combined with the high prevalence of subjects that already have cardiovascular complications at the time of diabetes diagnosis [4‐6], makes the screening for chronic complications highly recommended for all patients with type 2 diabetes (T2D). Early identification of complications can result in the prescription of more aggressive treatment regimen that have the potential to slow their natural progression. For example, a diagnosis of autonomic neuropathy by screening identifies patients at risk for adverse cardiovascular outcomes [7, 8]. Thus, results of complication screening can yield a better awareness of the interplay among micro- and microangiopathies [9, 10], and help clinicians to tailor treatment and the outpatient schedule visits [11]. In general, evidence of absence of complications or presence of multiple severe complications can both have important implications for appropriate management of diabetes and comorbidities [12]. Poor achievement of glucose targets [13] and the global under-utilization of glucose lowering drugs provided with cardiovascular protective effects [14, 15] suggests that a better screening of cardiorenal complications can promote improvements in therapeutic appropriateness for T2D.

However, the intensity and schedules of complication screening can vary considerably according to the healthcare setting (e.g. primary care versus specialist care) and availability of resources. Diagnostic workup for chronic diabetic complications should not be provided on demand based on patient’s complains, but based on routine schedules as recommended by evidence-based international standards [16].

There is mounting evidence of lower overall mortality in diabetic patients treated by a multi-disciplinary team within diabetes specialist outpatient services versus treatment in primary care [17]. However, it is still unknown whether, even within the same system of health care provision, screening intensity affects cardiovascular outcomes. We hypothesize that providing patients with an intensive session of multi-disciplinary screening of complications could result in better outcomes than when complications are screened in staggered sessions. Thus, the aim of this study was to evaluate whether an intensive screening of chronic complications resulted in long-term improvement of cardiovascular outcomes as compared to standard screening.

In this study, we show that T2D patients submitted to an intensive and comprehensive 1-day screening of complications had significantly better long-term cardiovascular outcomes than similar T2D patients who underwent routine screening at visits staggered during a year. To the best of our knowledge, this is the first report that the schedule of complication screening can affect cardiovascular outcomes of T2D patients.

Diabetes care in Italy is provided either by specialized diabetes outpatient clinics or by general practitioners. Diabetes clinics are provided with a multidisciplinary team of endocrinologists/diabetologists, trained nurses, podologists and psychologists. Close collaboration with general and vascular surgeons also allows the comprehensive management of the diabetic foot and peripheral arterial disease. Studies performed during the last 10 years reported that patients attending diabetes clinics had lower all-cause mortality rates than those treated by primary care physicians [17]. Though it is impossible to rule out all confounders, those studies highlight the benefits of a comprehensive management of diabetes. Results of our study indicate that, even within the same model of care, the way patients are involved in the screening program is associated with long-term cardiovascular outcome.

Patients receiving screening of micro- and macroangiopathy during a single day session had a 30% lower risk of experiencing a 4p-MACE over the following 6 years than patients whose complications were screened in separate sessions. We calculate that 22 patients (95% CI from 13 to 137) needed to be subjected to intensive screening to prevent one 4p-MACE in the subsequent 5 years. The benefit was particularly evident for the risk of HF, which was reduced by 57%. This result is particularly important since HF is a major cause of hospitalization in diabetic patients in Italy [22]. For this outcome, we estimated that 32 patients (95% CI from 23 to 107) were needed to be exposed to the intensive screening to avoid one hospitalization for HF in the subsequent 5 years. It should be highlighted that, despite in-hospital mortality after a MACE is typically increase in T2D [23], we found no difference between the two groups in the rates of fatal cardiovascular events, and all-cause mortality.

To understand drivers of the different cardiovascular outcome in the two groups, we analyzed the trends in cardiovascular risk factors and medications between baseline and censoring. We found only a nominally significant difference in systolic blood pressure, that may be of limited clinical relevance (1.1 mm Hg) and no significant between-group difference in HbA1c, BMI, eGFR, and lipid profile, nor in the modification of therapies over time. This finding suggests that one or more unmeasured factor(s) was driving the differential outcome between the two groups. We hypothesize that involvement in an intensive screening program improved patients’ awareness and positive attitudes towards diabetes and its complications. The benefit of increasing patients’ awareness have been recently reported in a large population-based cohort study, showing how empowerment programs for patients with T2D was associated with reduction in cardiovascular complications and overall mortality rate [24]. In addition, a complete snapshot of diabetes and complication burden may have driven further diagnostics and appropriate therapeutic tailoring that are not captured by comparing the average change in medications in the two groups. It is possible that, in the intensive screening group more than in the control group, the diagnostic and therapeutic approach was optimized especially in patients at higher cardiovascular risk. The optimized approach might also have led to higher adherence to prescribed treatment in the intensive group (e.g. for instance as suggested by the lower systolic blood pressure despite similar prescription therapy over follow-up) that is known to be associated with reduced risk of hospitalization and all-cause mortality in patients with T2D [25]. HbA1c variability, a well-known proxy of long-term adherence, is in fact associated with cardiovascular outcomes [26]. The intensive screening group, with a more comprehensive approach to the patients, might also have led to a healthier life style (e.g. level of physical activity and dietary pattern) that has been shown to be associated with reduced CV complications independently from established CVD risk factors [27]. However, due to the lack of systematic collection of this information in the EMR, we were unable to evaluate whether changes in lifestyle where influenced by the intensity screening visit.

Interpretation of our findings needs to carefully take into account limitations of this observational study. First, the screening strategy was not randomized, but based on patient’s and diabetologist’s preference. We tried to control for as many confounders as possible: PSM and multiple Cox regression analysis led to the same results, suggesting that known confounders were appropriately handled. It is however impossible to eliminate bias and adjust for unknown confounders due to unmeasured and unmeasurable variables.

We show that patients subjected to an intensive screening of diabetic complications followed by dedicated counseling had a significantly better cardiovascular outcome than patients subjected to standard staggered complication screening. In the absence of any change in cardiovascular risk factors and medications, such benefit may have resulted from improved patients’ compliance and attitudes towards diabetes management. Due to the possible role of unmeasured confounding factors, the observational design of this study does not allow to draw a firm causal relationship between the single-day intensive screening and the reduced incidence of non-fatal cardiovascular events. However, our results pave the way to the design new studies to validate these findings. Ideally, such studies should be randomized, e.g. by means of the design of a pragmatic clinical trial, in order to circumvent possible residual bias from unmeasured variables that cannot be excluded in observational studies.