Abstract
Background and Purpose
Infections have been found to increase the risk of stroke over the short-term. We hypothesized that stroke risk would be highest shortly after a sepsis hospitalization, but that the risk would decrease, yet remain up to 1-year after sepsis.
Methods
This case-crossover analysis utilized data obtained from the California State Inpatient Database of the Healthcare Cost and Utilization Project (HCUP). All stroke admissions were included. Exposure was defined as hospitalization for sepsis or septicemia 180, 90, 30 or 15 days before stroke (risk period) or similar time intervals exactly 1 or 2 years before stroke (control period). Conditional logistic regression was used to calculate the odds ratio and 95% confidence interval (OR, 95% CI) for the association between sepsis/septicemia and ischemic or hemorrhagic stroke.
Results
Ischemic (n=37,377) and hemorrhagic (n=12,817) strokes that occurred in 2009 were extracted where 3188 (8.5%) ischemic and 1101 (8.6%) hemorrhagic stroke patients had sepsis. Sepsis within 15 days prior to the stroke placed patients at the highest risk of ischemic (OR 28.36, 95% CI 20.02 –40.10) and hemorrhagic stroke (OR 12.10, 95% CI 7.54–19.42); however while the risk decreased, it remained elevated 181- 365 days after sepsis for ischemic (OR 2.59, 95%CI 2.20–3.06) and hemorrhagic (O 3.92, 95%CI 3.29–4.69) strokes. There was an interaction with age (p=0.0006); risk of developing an ischemic stroke within 180 days of hospitalization for sepsis increased 18% with each 10-year decrease in age.
Conclusion
Risk of stroke is high after sepsis, and this risk persists for up to a year. Younger sepsis patients have a particularly increased risk of stroke after sepsis.
Keywords: Acute Stroke, Stroke Outcomes
Stroke is the fifth leading cause of death in the US and the leading cause of serious long-term adult disability, with approximately 795,000 stroke events in the US each year.1 While much is known about long-term stroke risk factors, such as hypertension, diabetes, and atherosclerotic disease, much less is known about short-term risk factors, or triggers, for stroke.2
Infection has been identified as a potential risk factor and trigger for stroke. Chronic infection, as assessed by serologies against several common bacterial and viral infections, was associated with increased long-term stroke risk.3 In a case-crossover analysis from the Cardiovascular Health Study, a recent hospitalization for infection was associated with an increased risk of stroke.4
Recent evidence suggests that severe sepsis is associated with new-onset atrial fibrillation, thereby increasing risk of stroke.5 Furthermore, a population-based cohort study from Denmark showed that about 80% of cardiovascular events after exposure to bacteremia occurred during the index hospitalization.6 While other work shows that the risk of stroke is the highest in the first 3 to 15 days after infection.4, 7 We hypothesized that sepsis would be associated with risk of stroke after hospitalization for septicemia, and that the risk would be highest closest in time to the event.
Methods
Study Setting and Inclusion/Exclusion Criteria
This case-crossover analysis utilized data obtained from the California State Inpatient Database (SID), Healthcare Cost and Utilization Project (HCUP), Agency for Healthcare Research and Quality (AHRQ) for 2007–2009. Under HCUP, claims data for each discharge is collected, de-identified, and standardized from various states and then made available to researchers. The California SID contains data for all patients hospitalized in non-federal acute care California hospitals. Data elements include demographic information such as age, sex, race, and insurance payer. For each admission discharge diagnosis code (up to 25 ICD-9 codes), month of discharge, length of stay in hospital, and AHRQ comorbidity measures are available. There are designations for primary diagnosis and whether the condition was present on arrival for each ICD-9 code, allowing identification of pre-existing diagnoses versus complications that arise during hospitalization. A “visitlink” variable allows for tracking a patient over time through multiple hospital admissions.
The study population comprised patients with an ischemic or hemorrhagic stroke during hospitalization in any non-federal acute care hospital in California in the year 2009. The exposure of interest was septicemia or sepsis. A case-crossover analysis was used to investigate the association between septicemia/sepsis and subsequent stroke. This design is useful in studying acute events, such as stroke, brought on by exposures that transiently increase the risk for having an event.8 Data from a relatively short risk period preceding the event (“case” period) is compared to another “control” time period in the same individual, and exposures that are present more frequently in the risk period than the control period can be considered to be precipitants. In this study design cases act as their own controls, and thus the design inherently controls for inter-individual variability and confounding.4
Exposure and Covariates
The prior exposure was defined as hospitalization for sepsis 365, 180, 90, 30 or 15 days before stroke (case period) or similar time intervals exactly 1 or 2 years before stroke (control period). Septicemia or sepsis was defined by the following diagnostic codes present on arrival at any diagnostic position: 038.xx (septicemia), 020.0 (septicemic), 790.7(bacteremia), 117.9 (disseminated fungal infection), 112.5 (disseminated candida infection), 995.91(sepsis), 995.92 (severe sepsis) or 785.52 (septic shock).5 These diagnostic codes have been previously used in identification of sepsis cases in administrative datasets.5 Patients with admission codes for endocarditis and meningococcemia were not included in the sample as these are known direct risk factors for stroke. Septicemia or sepsis before stroke was considered as an exposure event. Risk time periods assessed were 0–15 days, 0–30 days, 0–90, 0–180 days and 0–365 days.6, 7 In addition to the risk time periods described above we assessed the risk of stroke within the time intervals 0–15 days, 16–30 days, 31–90 days, 91–180 days and 181–365 days.
Outcomes
Ischemic stroke was defined using ICD-9 codes 433.x1 (“x,” the fourth digit, can vary to specify a specific arterial distribution), 434 (excluding 434.x0), or 436 present at any diagnostic position between DX1 and DX12. Cases were excluded if any “traumatic brain injury” ICD-9-CM code (800 to 804, 850 to 854) or “rehabilitation care” ICD-9-CM code (V57) was present as the primary diagnosis.8 Hemorrhagic stroke was defined with ICD-9 codes 430–431 present at any diagnostic position.9
The California SID does not provide separable dates for each ICD-9 code within the same hospitalization, potentially limiting the assignment of temporal relationships among events within each admission. Simply deleting all hospital admissions in which stroke and sepsis occurred together, however, would lead to missing many cases of in-hospital stroke that were precipitated by sepsis. Thus two different datasets were created, one utilizing more restrictions on cases than the other. In both datasets we only included sepsis present on arrival, to avoid the inclusion of patients who developed sepsis as a complication of stroke (reverse causation). In the first dataset (the primary analytic sample), all hospital admissions in which both stroke and sepsis were present on arrival were omitted, to avoid uncertain temporality. The second dataset (the “restrictive” dataset) omitted all admissions for which any kind of stroke and sepsis occurred in the same admission during all three years of data. Thus, using the restrictive dataset analysis, all comparisons were only between separate admissions for stroke and sepsis. In addition, any sepsis that may have occurred in the prior years due to a stroke was also removed from the analysis in the second dataset. In the restrictive dataset, strokes that may have occurred due to hospitalization with sepsis were deleted if stroke was indicated to be present on admission.
Statistical analysis
Conditional logistic regression stratified on the variable visitlink, a variable created within the database to link patients in the sample without identifying information, was used to compute odds ratios (ORs) and 95% confidence intervals (95% CI) for any hospital admission with stroke within 0–15 days, 0–30 days, 0–180 and 0–365 days after exposure. We further assessed risk of stroke post sepsis at non-overlapping time intervals after sepsis by investigating the risk at 0–15 days, 16–30 days, 31–90 days, 91–180 days and 181–365 days. Ischemic and hemorrhagic stroke were studied separately for each time period. Interactions between sepsis and age, sex and diabetes were investigated, and stratification by age performed as indicated. All hypothesis tests performed during the analysis of the primary endpoints are two-sided and use an alpha of 0.05.
Results
Primary Analysis Study Population
A total of 37,377 ischemic strokes and 12,817 hemorrhagic strokes that occurred in 2009 were extracted in the primary dataset. Of the ischemic strokes, 3188 (8.5%) had at least 1 case of sepsis in the 365-day risk period prior to their stroke. Among hemorrhagic strokes 1,101 (8.6%) had at least 1 case of sepsis in the 365-day risk period prior to their stroke.
Mean hospital length of stay was higher in patients with both stroke and sepsis (ischemic stroke 9.7 ± 16.8 days, hemorrhagic stroke 13.2 ± 19.7 days) as compared to patients with stroke only (ischemic 4.9 ± 7.9 days, hemorrhagic 8.2 days ± 11.9 days; Table 1). The frequencies of comorbidities such as diabetes, chronic obstructive pulmonary disease (COPD), paralysis, and renal failure were higher in patients with both stroke and sepsis or septicemia when compared to patients with stroke only (Table 1). The higher prevalence of paralysis among those with sepsis and stroke could reflect prior stroke, or that patients with sepsis are more likely to have embolic strokes or severe strokes that are more likely to involve paralysis. It is difficult to assess these possibilities because we did not have the granular detail that would allow us to assess stroke severity or etiologic from these data.
Table 1.
Baseline characteristics of ischemic and hemorrhagic stroke patients with and without sepsis/septicemia preceding stroke
| Variable | All Ischemic stroke cases (n= 37377) |
Ischemic stroke cases without exposure to septicemia (n= 34189) |
Ischemic stroke cases with exposure to septicemia (n = 3188) |
All Hemorrhagic stroke cases (n= 12817) |
Hemorrhagic stroke cases without exposure to sepsis (n= 11716) |
Hemorrhagic stroke cases with exposure to sepsis (n=1101) |
|
|---|---|---|---|---|---|---|---|
| Age (years) | 71.7 | 71.5 | 74.1 | 66.7 | 66.4 | 69.3 | |
| standard deviation(range) | 14.6(0–104) | 14.6(0–104) | 13.6(0–101) | 16.9(0–101) | 17(0–101) | 15.8(0–97) | |
| Number of chronic conditions | 6.8 | 6.7 | 8.3 | 6 | 5.8 | 7.8 | |
| standard deviation(range) | 2.9(1–22) | 2.8(1–22) | 3.1(1–21) | 2.9(1–101) | 2.8(1–20) | 3.1(1–18) | |
| Length of stay(days) | 5.4 | 4.9 | 9.7 | 8.6 | 8.2 | 13.2 | |
| standard deviation(range) | 9.1(0–351) | 7.9(0–345) | 16.8(0–351) | 12.8(0–274) | 11.9(0–253) | 19.7(0–274) | |
| Women, No. (%) | 19476(52.2) | 17798(52.2) | 1678(52.8) | 6123(48.2) | 5626(48.4) | 497(45.3) | |
| Race No. (%) | |||||||
| White | 21824(60.87) | 20081(61.2) | 1743(56.9) | 6443(53.4) | 5956(54) | 487(46.5) | |
| Black | 3601(10.4) | 3239(9.9) | 362(11.8) | 1030(8.5) | 913(8.3) | 117(11.2) | |
| Hispanic | 6528(18.2) | 5900(18) | 628(20.5) | 2708(22.4) | 2436(22) | 272(26) | |
| Asian/pacific Islander | 3237(9) | 2956(9) | 281(9.2) | 1615(13.4) | 1458(13.2) | 157(15) | |
| Other | 640(1.76) | 591(1.8) | 49(1.6) | 269(2.23) | 255(2.3) | 14(1.3) | |
| Disposition of patient at discharge No. (%) | |||||||
| Died in hospital | 2098 (5.62) | 1788 (5.2) | 310 (9.7) | 2732 (21.3) | 2528 (21.6) | 204 (18.5) | |
| AHRQ Comorbidity measures No. (%) | |||||||
| Valvular disease | 3236(8.7) | 2896(8.5) | 340(10.7) | 660(5.15) | 586(5) | 74(6.7) | |
| Metastatic cancer | 560(1.5) | 489(1.4) | 71(2.2) | 341(2.7) | 296(2.5) | 45(4) | |
| Renal failure | 5502(14.7) | 4556(13.3) | 946(29.7) | 1528(11.9) | 1224(10.4) | 304(27.6) | |
| Congestive heart failure | 5332(14.3) | 4521(13.2) | 811(25.4) | 1229(9.6) | 1012(8.62) | 217(19.7) | |
| Chronic pulmonary disease | 5338(14.3) | 4661(13.6) | 677(21.2) | 1547(12) | 1344(11.5) | 203(18.4) | |
| Coagulopathy | 1128(3) | 888(2.6) | 240(7.5) | 995(7.8) | 822(7) | 173(15.7) | |
| Psychoses | 1428(3.8) | 1253(3.6) | 175(4.5) | 455(3.5) | 401(3.4) | 54(4.9) | |
| Peripheral vascular disorders | 3385(9) | 2979(8.7) | 406(12.7) | 688(5.4) | 577(4.9) | 111(10) | |
| Paralysis | 1882(5.04) | 1580(4.6) | 302(9.5) | 897(7) | 769(6.6) | 128(11.6) | |
| Other neurological disorders | 718(1.9) | 489(1.4) | 229(7.2) | 528(4.1) | 412(3.5) | 116(10.5) | |
| Hypertension (combine uncomplicated and complicated) |
29982(80.2) | 27450(80.3) | 2532(79.4) | 9357(73) | 8509(72.6) | 848(77) | |
| Drug abuse | 966(2.6) | 897(2.6) | 69(2.2) | 554(4.2) | 505(4.3) | 39(3.5) | |
| Diabetes with chronic complications |
3278(8.8) | 2798(8.2) | 480(15) | 654(5.1) | 530(4.5) | 124(11.3) | |
| Diabetes, uncomplicated | 9477(25.4) | 8566(25) | 911(28.6) | 2636(20.6) | 2342(20) | 294(26.7) | |
| Rheumatoid arthritis/collagen vascular diseases |
897(2.4) | 788(2.3) | 109(3.4) | 247(1.9) | 219(1.9) | 28(2.54) | |
| Alcohol abuse | 1510(4) | 1399(4) | 111(3.5) | 864(6.7) | 791(6.8) | 73(6.6) | |
Sepsis was associated with an increased risk of ischemic (OR 28.36, 95% CI 20.02 –40.10) and hemorrhagic (OR 12.10, 95% CI 7.54–19.42) stroke within 15 days. The risk of stroke after sepsis persisted as the time interval from sepsis to stroke increased, but the magnitude of the risk decreased as the time interval was increased (Table 2). The risk of ischemic (OR 3.92, 95% CI 3.58–4.29) and hemorrhagic stroke (OR 7.38, 95% CI 6.54–8.34) remained significantly elevated after increasing the time interval to 365 days after the sepsis event.
Table 2.
Cumulative Association of hospitalization for sepsis or septicemia with risk of ischemic and hemorrhagic stroke, based on case-crossover analysis (using dataset 1**)
| Dataset 1 | Ischemic Stroke | Hemorrhagic Stoke |
|---|---|---|
| OR (95% CI) | OR (95% CI) | |
| Hospitalization for sepsis/septicemia within 15 d before stroke |
28.4 (20.0 –40.1) | 12.1 (7.54–19.4) |
| Hospitalization for sepsis/septicemia within 30 d before stroke |
13.9 (10.9–17.5) | 16.0 (11.8–21.8) |
| Hospitalization for sepsis/septicemia within 90 d before stroke |
6.46 (5.56–7.50) | 12.3 (9.34–15.2) |
| Hospitalization for sepsis/septicemia within 180 d before stroke |
4.97 (4.40–5.61) | 9.78 (8.34–11.5) |
| Hospitalization for sepsis/septicemia within 365 d before stroke |
3.92 (3.58–4.29) | 7.38 (6.54–8.34) |
OR indicates odds ratio; CI, confidence interval.
All p-values were < 0.0001
Dataset 1: In the first dataset all hospital admissions in which both stroke and sepsis were present on arrival were deleted.
In analyses utilizing non-overlapping time intervals from sepsis to stroke (Table 3), the risk of ischemic stroke was greatest in the first 15 days post sepsis hospitalization, decreased markedly from 16–30 days after sepsis, but remained increased by about 2.5-fold out as long as 181–365 days after sepsis. The pattern for hemorrhagic stroke was somewhat different, with an elevated risk in the first 15 days, with the risk decreasing more linearly for time intervals 16–30, 31–90 days, 91–180 days and 181–365 days (Table 3). The risk remained approximately 4 times higher even in the 181–365 day interval.
Table 3.
Association of hospitalization for sepsis or septicemia in non-overlapping time intervals with risk of ischemic and hemorrhagic stroke, based on case-crossover analysis (using dataset 1**)
| Dataset 1 | Ischemic Stroke | Hemorrhagic Stroke | ||
|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | |
| Hospitalization for sepsis/septicemia 0–15 days before stroke |
28.4 | 20.0–40.1 | 12.1 | 7.54–19.4 |
| Hospitalization for sepsis/septicemia 16–30 days before stroke |
4.00 | 2.86–5.60 | 8.42 | 5.82–12.2 |
| Hospitalization for sepsis/septicemia 31–90 days before stroke |
2.53 | 2.07–3.09 | 5.48 | 4.21–7.13 |
| Hospitalization for sepsis/septicemia 91–180 days before stroke |
2.46 | 2.03–2.99 | 4.79 | 3.87–5.92 |
| Hospitalization for sepsis/septicemia 181–365 days before stroke |
2.59 | 2.20–3.06 | 3.92 | 3.29–4.69 |
OR indicates odds ratio; CI, confidence interval.
All p-values were < 0.0001
Dataset 1: In the first dataset all hospital admissions in which both stroke and sepsis were present on arrival were deleted.
Secondary Analysis
A total of 37,377 ischemic strokes and 12,817 hemorrhagic strokes that occurred in 2009 were in the secondary restrictive dataset. Of the ischemic strokes, 2301 (6.2%) had at least 1 case of sepsis in the 365-day risk period prior to their stroke. Among hemorrhagic strokes 879 (6.9%) had at least 1 case of sepsis in the 365-day risk period prior to their stroke.
Hospitalization with sepsis within 15 days was associated with an elevated risk of ischemic (OR 5.28, 95%CI 3.65–7.64) and hemorrhagic stroke (OR 3.45, 95%CI 2.52–5.54). The risk of stroke after sepsis persisted after increasing the time interval to 365 days after the sepsis event, and the magnitude of risk decreased as the time interval increased (Table 4). The risk of ischemic (OR 3.86, 95%CI 3.59–4.15) and hemorrhagic stroke (OR 4.14, 95%CI 3.68–4.67) remained elevated in the interval up to 365 days after the sepsis event.
Table 4.
Cumulative Association of hospitalization for sepsis or septicemia with risk of ischemic and hemorrhagic stroke, based on case-crossover analysis (using dataset 2**)
| Dataset 2 | Ischemic Stroke | Hemorrhagic Stoke |
|---|---|---|
| OR (95% CI) | OR (95% CI) | |
| Hospitalization for sepsis/septicemia within 15 d before stroke |
5.28 (3.65 – 7.64) | 3.45 (2.04–5.84) |
| Hospitalization for sepsis/septicemia within 30 d before stroke |
6.35 (5.07–7.94) | 4.69 (3.32–6.62) |
| Hospitalization for sepsis/septicemia within 90 d before stroke |
4.78 (4.17–5.49) | 4.60 (3.68–5.76) |
| Hospitalization for sepsis/septicemia within 180 d before stroke |
4.23 (3.85–4.67) | 4.53 (3.86–5.32) |
| Hospitalization for sepsis/septicemia within 365 d before stroke |
3.86 (3.59–4.15) | 4.14 (3.68–4.67) |
OR indicates odds ratio; CI, confidence interval.
All p-values were < 0.0001
The second dataset deleted all admissions in which any kind of stroke and sepsis occurred in the same admission from all three years of data. Thus the comparison was only between separate admissions for stroke and sepsis.
In analyses of non-overlapping time periods (Table 5), risk of ischemic stroke was similar, ranging from approximately 4-fold to 6-fold, throughout all time periods. We found the risk of hemorrhagic stroke was also similar across time intervals, with the risk slightly decreasing with each increase in time interval from sepsis event (Table 5).
Table 5.
Association of hospitalization for sepsis or septicemia at each specific time interval with risk of ischemic and hemorrhagic stroke, based on case-crossover analysis (using dataset 2**)
| Ischemic | Hemorrhagic | |||
|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | |
| Hospitalization for sepsis/septicemia 0–15 days before stroke | 5.28 | 3.65–7.63 | 3.45 | 2.04–5.84 |
| Hospitalization for sepsis/septicemia 15–30 days before stroke |
4.58 | 3.56–5.89 | 3.74 | 2.52 – 5.54 |
| Hospitalization for sepsis/septicemia 30–90 days before stroke |
3.34 | 2.85–3.91 | 3.63 | 2.77–4.76 |
| Hospitalization for sepsis/septicemia 90–180 days before stroke |
3.14 | 2.76–3.57 | 3.62 | 2.93–4.48 |
| Hospitalization for sepsis/septicemia 180–365 days before stroke |
2.94 | 2.65–3.26 | 3.25 | 2.73–3.86 |
OR indicates odds ratio; CI, confidence interval.
All p-values were < 0.0001
The second dataset deleted all admissions in which any kind of stroke and sepsis occurred in the same admission from all three years of data. Thus the comparison was only between separate admissions for stroke and sepsis.
Interactions with Age, Sex and Diabetes
The risk of having an ischemic stroke within 180 days of hospitalization for sepsis varied significantly with age (p-value for interaction = 0.0006). The change in the OR for the association of sepsis with stroke risk increased 18% with each 10-year decrease in age (OR per 10 year age decrease 1.18, 95%CI 1.05–1.45). No significant interaction with age was seen in the hemorrhagic stroke dataset. We further investigated this interaction by stratifying the dataset into age categories (<45, 45–65 and >65 years). These groups were chosen based on prior age categories used for stroke. The risk for those <45 years was highest (OR for stroke associated with sepsis 6.00, 95% CI 1.62–22.1), followed by similar risk profiles for those aged 45–65 (OR 2.57, 95%CI 1.71–3.85) and greater than 65 (OR 2.32, 95%CI 1.86–2.91).
The risk of ischemic stroke within 180 days of hospitalization for sepsis also varied significantly with diabetes status (p = 0.044). Individuals with diabetes had a slightly lower risk (OR 4.11, 95% CI 3.59–4.70)) of ischemic stroke within 180 days of hospitalization for sepsis as compared to individuals without diabetes (OR 5.39, 95% CI 5.00–5.82). There was no significant interaction with diabetes status in the hemorrhagic stroke dataset (p=0.3473).
There was no interaction with sex (p=0.440).
Discussion
We found that sepsis increased the risk of stroke as long as 365 days after an admission with sepsis. The risk of stroke was highest within 15 days after sepsis, with the risk of stroke decreasing as the time from sepsis increased. The magnitude of the association still remained as high as a 3-fold increase in odds of a stroke at 180 days for both ischemic and hemorrhagic stroke. We also found that the magnitude of association was greater for younger patients.
These findings are consistent with studies investigating the association of sepsis on stroke in Taiwan, though in our US population the relationship was stronger with ischemic stroke than with hemorrhagic stroke.9, 10 Associations between sepsis and vascular disease are plausible given the role of endothelial dysfunction in sepsis pathophysiology.11 Sepsis leads to systemic inflammation, hemodynamic dysfunction and collapse, and coagulopathy, thereby increasing the risk for stroke.12, 13 Chronic medical conditions, including obesity, diabetes, heart disease, and smoking, have also been associated with chronic inflammation and subsequent increased risk of developing future sepsis, in addition to being risk factors for stroke.14–20 These conditions could be a source of confounding, since the risk factors could be associated with both sepsis and stroke risk; however, by using the case-crossover analysis we minimized inter-individual confounding. Moreover, inflammatory biomarkers measured at a healthy baseline have been linked to sepsis risk in the REGARDS cohort, and have been linked to risk of stroke in other cohorts.21–27 Another study identified a potential mechanism through sepsis-induced atrial fibrillation.4–7, 9 Furthermore, serological evidence of infection was associated with a modestly increased risk of incident stroke in population-based studies, with common bacterial and herpesvirus infections associated with increased stroke risk.4 Our study demonstrates that there is an even greater increased risk of stroke after a more severe infectious event such as sepsis, indicating a potential dose-response relationship with time from infectious or inflammatory burden and risk of stroke. It is unclear if the increased risk of stroke after sepsis is due to the shared risk factors and comorbid conditions that place a patient at risk for both sepsis and stroke, or if sepsis is independently associated with risk of stroke. Studies on sepsis mortality indicate that sepsis patients are at increased risk of mortality post-sepsis, and that 70% of the deaths were due to cardiovascular or pulmonary disease.28 These findings support our hypothesis that among the sepsis patients who survive their sepsis hospitalization, they are at increased risk of cardiovascular disease and stroke, and subsequent mortality due to their disease. Whether this is due to sepsis susceptibility or post-sepsis inflammatory effects is unknown.
Additionally, we found that younger patients were at a further increased risk of stroke after sepsis compared to older patients; for every decade younger age, the risk increased by almost 20% for ischemic, but not hemorrhagic, stroke after sepsis. These findings are similar to the findings of Lee et al, but we did not see an interaction with age and hemorrhagic stroke, only with age and ischemic stroke.9 This could be due to the differences in the underlying populations between the two studies (Taiwan versus US), as well as the fact that the cases and controls in their study were selected based on exposure, rather than outcome, which would tend to inflate their estimates of the association.
While the risk of stroke is greater in older adults, approximately 10–14% of all strokes occur in people 18–45 years old.29–35 Conventional risk factors such as hypertension, diabetes, and smoking, moreover, may not fully account for the risk of stroke in patients aged 18–45.36 The increasing prevalence of stroke in the young, coupled with greater heterogeneity in stroke etiology within the younger age group than in the older stroke population, presents a unique and vulnerable patient population where risk reduction efforts are of increasing importance.
The interaction with diabetes reducing the risk of stroke after sepsis is different than what has been reported in the literature. There was not an interaction with diabetes in the CHS study, but there was an interaction with IMT as a measure of atherosclerosis.37 Furthermore, while the risk in general is greatest shortly after sepsis, the risk does remain elevated out to 365 days post-sepsis. This could indicate that the diseases co-vary, which is supported by the data on the number of chronic conditions, or that these patients are sicker patients in general. Another explanation could be a long-term biological effect from the sepsis event. Other studies have shown long-term effects of pneumonia on risk of stroke, and considering the toll sepsis takes on the inflammatory system this effect could be more pronounced after sepsis than it is after pneumonia.37 Furthermore, we do not have information on baseline inflammation and risk of stroke. We do not know if the risk returns to a pre-sepsis baseline or if the elevated risk remains during the life course after sepsis. If the risk of stroke remains elevated post-sepsis this could be due to the inflammation itself, or it could reflect the co-morbidities and complications associated with hospitalization for sepsis.
Our study has limitations. Prospective case ascertainment and detailed clinical information were not available. For example, we did not have data on specific infectious etiologies of sepsis. We relied on administrative data diagnosis codes, and information on stroke severity was not available. The exclusion criteria could have excluded stroke cases that were more severe, as stroke severity is associated with risk of sepsis during stroke hospitalization. The co-morbidities were captured through the AHRQ co-morbidity measures, which are based on discharge diagnoses. This could underreport the co-morbidities commonly associated with stroke. We were only able to capture patients who were hospitalized, and in the state of California. Patients who had hospitalizations in other states, or who died of an event prior to being admitted to a hospital, were not captured. The case-crossover design also does not account for the increased risk associated with the aging of the patient over time and their concomitant development of new risk factors. To minimize these concerns, however, we limited the control time windows to the prior 2 years. Another limitation is the potential for immortal time bias, as the extended risk of stroke over time could be influenced by the high long-term mortality associated with sepsis. The discrepancy between the proportions of ischemic and hemorrhagic stroke in our study (66% ischemic) when compared to all strokes in the US (87% ischemic) could be due to the exclusions we applied for this study, as well as the fact that ischemic stroke hospitalizations are more likely not to be hospitalized.
Our study has strengths, as well, however. By utilizing the case-crossover design, in which each patient served as his or her own control, we eliminated inter-individual variability. The use of a large administrative database increases the generalizability of the study. Moreover, our very large sample size, allowed us to detect interactions with age and diabetes, and to study associations among populations of patients that are often difficult to capture, including young and hemorrhagic stroke patients.
This study suggests that sepsis patients remain at risk for stroke as long as 365 days from their sepsis hospitalization, with the risk being greatest shortly after the sepsis admission. Younger sepsis patients have an increased risk of stroke after sepsis admission. This study identifies a unique group of patients at increased risk for stroke. Future studies are needed to confirm these relationships in other patient populations, determine if and when the risk returns to a baseline level of risk, investigate mechanisms of the increased risk of stroke after sepsis, and determine effective strategies to reduce this risk.
Acknowledgments
None
Funding Sources
Dr. Boehme is supported by National Institute of Neurological Disorders and Stroke (NINDS) National Institute of Health (NIH) T32 NS007153-31 and L30 NS093600 grants. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NINDS or the NIH.
Disclosures
Dr. Elkind receives compensation for providing consultative services for Biotelemetry/Cardionet, BMS-Pfizer Partnership, Boehringer-Ingelheim, and Sanofi-Regeneron; serves on the National, Founders Affiliate, and New York City chapter boards of the American Heart Association/American Stroke Association; and receives royalties from UpToDate for chapters related to cryptogenic stroke.
References
- 1.Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics--2015 update: a report from the American Heart Association. Circulation. 2015;131:e29–e322. doi: 10.1161/CIR.0000000000000152. [DOI] [PubMed] [Google Scholar]
- 2.Elkind MS. Why now? Moving from stroke risk factors to stroke triggers. Curr Opin Neurol. 2007;20:51–57. doi: 10.1097/WCO.0b013e328012da75. [DOI] [PubMed] [Google Scholar]
- 3.Elkind MS, Ramakrishnan P, Moon YP, et al. Infectious burden and risk of stroke: the northern Manhattan study. Archives of neurology. 2010;67:33–38. doi: 10.1001/archneurol.2009.271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Elkind MS, Carty CL, O'Meara ES, et al. Hospitalization for infection and risk of acute ischemic stroke: the Cardiovascular Health Study. Stroke; a journal of cerebral circulation. 2011;42:1851–1856. doi: 10.1161/STROKEAHA.110.608588. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Walkey AJ, Hammill BG, Curtis LH, Benjamin EJ. Long-term outcomes following development of new-onset atrial fibrillation during sepsis. Chest. 2014;146:1187–1195. doi: 10.1378/chest.14-0003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Dalager-Pedersen M, Sogaard M, Schonheyder HC, Nielsen H, Thomsen RW. Risk for myocardial infarction and stroke after community-acquired bacteremia: a 20-year population-based cohort study. Circulation. 2014;129:1387–1396. doi: 10.1161/CIRCULATIONAHA.113.006699. [DOI] [PubMed] [Google Scholar]
- 7.Clayton TC, Thompson M, Meade TW. Recent respiratory infection and risk of cardiovascular disease: case-control study through a general practice database. Eur Heart J. 2008;29:96–103. doi: 10.1093/eurheartj/ehm516. [DOI] [PubMed] [Google Scholar]
- 8.Maclure M. The case-crossover design: a method for studying transient effects on the risk of acute events. Am J Epidemiol. 1991;133:144–153. doi: 10.1093/oxfordjournals.aje.a115853. [DOI] [PubMed] [Google Scholar]
- 9.Lee JT, Chung WT, Lin JD, et al. Increased risk of stroke after septicaemia: a population-based longitudinal study in Taiwan. PLoS One. 2014;9:e89386. doi: 10.1371/journal.pone.0089386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ou SM, Chu H, Chao PW, et al. Long-Term Mortality and Major Adverse Cardiovascular Events in Sepsis Survivors. A Nationwide Population-based Study. American journal of respiratory and critical care medicine. 2016;194:209–217. doi: 10.1164/rccm.201510-2023OC. [DOI] [PubMed] [Google Scholar]
- 11.Aird WC. Endothelium as a therapeutic target in sepsis. Current drug targets. 2007;8:501–507. doi: 10.2174/138945007780362782. [DOI] [PubMed] [Google Scholar]
- 12.Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med. 2001;29:1303–1310. doi: 10.1097/00003246-200107000-00002. [DOI] [PubMed] [Google Scholar]
- 13.Kellum JA, Kong L, Fink MP, et al. Understanding the inflammatory cytokine response in pneumonia and sepsis: results of the Genetic and Inflammatory Markers of Sepsis (GenIMS) Study. Archives of internal medicine. 2007;167:1655–1663. doi: 10.1001/archinte.167.15.1655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Wang HE, Shapiro NI, Griffin R, Safford MM, Judd S, Howard G. Chronic medical conditions and risk of sepsis. PLoS One. 2012;7:e48307. doi: 10.1371/journal.pone.0048307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cave MC, Hurt RT, Frazier TH, et al. Obesity, inflammation, and the potential application of pharmaconutrition. Nutr Clin Pract. 2008;23:16–34. doi: 10.1177/011542650802300116. [DOI] [PubMed] [Google Scholar]
- 16.Mathieu P, Poirier P, Pibarot P, Lemieux I, Despres JP. Visceral obesity: the link among inflammation, hypertension, and cardiovascular disease. Hypertension. 2009;53:577–584. doi: 10.1161/HYPERTENSIONAHA.108.110320. [DOI] [PubMed] [Google Scholar]
- 17.Tousoulis D. Inflammation in atherosclerosis: current therapeutic approaches. Curr Pharm Des. 2011;17:4087–4088. doi: 10.2174/138161211798764762. [DOI] [PubMed] [Google Scholar]
- 18.Brooks-Worrell B, Palmer JP. Immunology in the Clinic Review Series; focus on metabolic diseases: development of islet autoimmune disease in type 2 diabetes patients: potential sequelae of chronic inflammation. Clin Exp Immunol. 2012;167:40–46. doi: 10.1111/j.1365-2249.2011.04501.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rosner SA, Ridker PM, Zee RY, Cook NR. Interaction between inflammation-related gene polymorphisms and cigarette smoking on the risk of myocardial infarction in the Physician's Health Study. Hum Genet. 2005;118:287–294. doi: 10.1007/s00439-005-0052-6. [DOI] [PubMed] [Google Scholar]
- 20.Arnson Y, Shoenfeld Y, Amital H. Effects of tobacco smoke on immunity, inflammation and autoimmunity. Journal of autoimmunity. 2010;34:J258–J265. doi: 10.1016/j.jaut.2009.12.003. [DOI] [PubMed] [Google Scholar]
- 21.Chamorro A, Amaro S, Vargas M, et al. Interleukin 10, monocytes and increased risk of early infection in ischaemic stroke. Journal of neurology, neurosurgery, and psychiatry. 2006;77:1279–1281. doi: 10.1136/jnnp.2006.100800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Haeusler KG, Schmidt WU, Fohring F, et al. Cellular immunodepression preceding infectious complications after acute ischemic stroke in humans. Cerebrovascular diseases. 2008;25:50–58. doi: 10.1159/000111499. [DOI] [PubMed] [Google Scholar]
- 23.Emsley HC, Hopkins SJ. Acute ischaemic stroke and infection: recent and emerging concepts. The Lancet Neurology. 2008;7:341–353. doi: 10.1016/S1474-4422(08)70061-9. [DOI] [PubMed] [Google Scholar]
- 24.Emsley HC, Smith CJ, Hopkins SJ. Infection and brain-induced immunodepression after acute ischemic stroke. Stroke; a journal of cerebral circulation. 2008;39:e7. doi: 10.1161/STROKEAHA.107.500447. author reply e8. [DOI] [PubMed] [Google Scholar]
- 25.Vogelgesang A, Grunwald U, Langner S, et al. Analysis of lymphocyte subsets in patients with stroke and their influence on infection after stroke. Stroke; a journal of cerebral circulation. 2008;39:237–241. doi: 10.1161/STROKEAHA.107.493635. [DOI] [PubMed] [Google Scholar]
- 26.Dirnagl U, Klehmet J, Braun JS, et al. Stroke-induced immunodepression: experimental evidence and clinical relevance. Stroke; a journal of cerebral circulation. 2007;38:770–773. doi: 10.1161/01.STR.0000251441.89665.bc. [DOI] [PubMed] [Google Scholar]
- 27.Meisel C, Schwab JM, Prass K, Meisel A, Dirnagl U. Central nervous system injury-induced immune deficiency syndrome. Nature reviews Neuroscience. 2005;6:775–786. doi: 10.1038/nrn1765. [DOI] [PubMed] [Google Scholar]
- 28.Wang HE, Szychowski JM, Griffin R, Safford MM, Shapiro NI, Howard G. Long-term mortality after community-acquired sepsis: a longitudinal population-based cohort study. BMJ open. 2014;4:e004283. doi: 10.1136/bmjopen-2013-004283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Putaala J, Metso AJ, Metso TM, et al. Analysis of 1008 consecutive patients aged 15 to 49 with first-ever ischemic stroke: the Helsinki young stroke registry. Stroke; a journal of cerebral circulation. 2009;40:1195–1203. doi: 10.1161/STROKEAHA.108.529883. [DOI] [PubMed] [Google Scholar]
- 30.Adams HP, Jr, Kappelle LJ, Biller J, et al. Ischemic stroke in young adults. Experience in 329 patients enrolled in the Iowa Registry of stroke in young adults. Archives of neurology. 1995;52:491–495. doi: 10.1001/archneur.1995.00540290081021. [DOI] [PubMed] [Google Scholar]
- 31.Kittner SJ, Stern BJ, Wozniak M, et al. Cerebral infarction in young adults: the Baltimore-Washington Cooperative Young Stroke Study. Neurology. 1998;50:890–894. doi: 10.1212/wnl.50.4.890. [DOI] [PubMed] [Google Scholar]
- 32.Jacobs BS, Boden-Albala B, Lin IF, Sacco RL. Stroke in the young in the northern Manhattan stroke study. Stroke; a journal of cerebral circulation. 2002;33:2789–2793. doi: 10.1161/01.str.0000038988.64376.3a. [DOI] [PubMed] [Google Scholar]
- 33.Qureshi AI, Safdar K, Patel M, Janssen RS, Frankel MR. Stroke in young black patients. Risk factors, subtypes, and prognosis. Stroke; a journal of cerebral circulation. 1995;26:1995–1998. doi: 10.1161/01.str.26.11.1995. [DOI] [PubMed] [Google Scholar]
- 34.Naess H, Nyland HI, Thomassen L, Aarseth J, Nyland G, Myhr KM. Incidence and short-term outcome of cerebral infarction in young adults in western Norway. Stroke; a journal of cerebral circulation. 2002;33:2105–2108. doi: 10.1161/01.str.0000023888.43488.10. [DOI] [PubMed] [Google Scholar]
- 35.George MG, Tong X, Kuklina EV, Labarthe DR. Trends in stroke hospitalizations and associated risk factors among children and young adults, 1995–2008. Ann Neurol. 2011;70:713–721. doi: 10.1002/ana.22539. [DOI] [PubMed] [Google Scholar]
- 36.Ji R, Schwamm LH, Pervez MA, Singhal AB. Ischemic stroke and transient ischemic attack in young adults: risk factors, diagnostic yield, neuroimaging, and thrombolysis. JAMA neurology. 2013;70:51–57. doi: 10.1001/jamaneurol.2013.575. [DOI] [PubMed] [Google Scholar]
- 37.Corrales-Medina VF, Alvarez KN, Weissfeld LA, et al. Association between hospitalization for pneumonia and subsequent risk of cardiovascular disease. JAMA. 2015;313:264–274. doi: 10.1001/jama.2014.18229. [DOI] [PMC free article] [PubMed] [Google Scholar]