Serum IgG titers against periodontal pathogens are associated with cerebral hemorrhage growth and 3-month outcome.

To assess the influence of periodontal disease on cerebral hemorrhage and its clinical course, we examined the association of the serum IgG titer of periodontal pathogens with hemorrhage growth and 3-month outcome. We consecutively enrolled 115 patients with acute cerebral hemorrhage (44 females, aged 71.3 ± 13.1 years) and used ELISA to evaluate the serum IgG titers of 9 periodontal pathogens: Porphyromonas gingivalis, Aggregatibacter (A.) actinomycetemcomitans, Prevotella intermedia, Prevotella nigrescens, Fusobacterium (F.) nucleatum, Treponema denticola, Tannerella forsythensis, Campylobacter rectus, and Eikenella corrodens. Significant hematoma growth was defined as an increase in the volume of >33% or an absolute increase in the volume of >12.5 mL. A poor outcome was defined as a 3 or higher on the modified Rankin Scale. We observed hemorrhage growth in 13 patients (11.3%). Multivariate analysis revealed that increased IgG titers of A. actinomycetemcomitans independently predicted the elevated hemorrhage growth (odds ratio 5.26, 95% confidence interval 1.52-18.25, p = 0.01). Notably, augmented IgG titers of F. nucleatum but not A. actinomycetemcomitans led to a poorer 3-month outcome (odds ratio 7.86, 95% confidence interval 1.08-57.08, p = 0.04). Thus, we demonstrate that elevated serum IgG titers of A. actinomycetemcomitans are an independent factor for predicting cerebral hemorrhage growth and that high serum IgG titers of F. nucleatum may predict a poor outcome in patients with this disease. Together, these novel data reveal how systemic periodontal pathogens may affect stroke patients, and, should, therefore, be taken into consideration in the management and treatment of these individuals.

Introduction

Periodontal disease is a risk factor for diabetes mellitus, rheumatoid arthritis, and several types of cancer [1-3]. It is a chronic inflammatory disease caused by an immune response to periodontal bacteria and characterized by loss of connective tissue and alveolar bone support, ultimately causing tooth loss [4]. Meta-analysis from previous studies showed that the risk of stroke was significantly increased in individuals with periodontitis in which the relative risk was 1.63 (95% confidence interval [CI] 1.25–2.00) [5]. Furthermore, a large cohort study has demonstrated that periodontal disease is associated with the incidence of cardioembolic and atherothrombotic stroke, and that regular dental care might decrease stroke risk [6]. Periodontitis is related to an increase in systemic inflammation markers through exposure to Gram-negative bacteria, which are implicated in the etiology of stroke [7]. While the treatment of stroke has improved remarkably, the management of periodontal disease may not only improve the clinical course in patients but could help prevent stroke altogether. However, in these studies periodontal diseases were diagnosed by only oral examination.

It was reported that serum IgG titers of a certain periodontal pathogen are considered to reflect its periodontal status [8]. Recently, by analyzing serum antibody titers, certain periodontal pathogens have been identified as risk factors for systemic diseases, such as ischemic stroke, coronary heart disease, non-alcoholic fatty liver disease, and Alzheimer's disease [9-12]. Especially regarding ischemic stroke, the serum antibody level of Prevotella (P.) intermedia was significantly higher in atherothrombotic stroke patients than in patients with no previous stroke [9]. However, there is no reported association between the serum IgG titers of periodontal pathogens and cerebral hemorrhage. While it was reported that serum antibody titers of Fusobacterium (F.) nucleatum are a predictor of unfavorable outcome after stroke [13], a thorough investigation has yet to be conducted that identifies the types of pathogens associated with stroke and its clinical course. To determine if periodontal disease affects cerebral hemorrhage, we examined the relationship among serum IgG titers of periodontal pathogens, cerebral hemorrhage growth, and a 3-month outcome.

Materials and methods

Patients

Consecutive acute cerebral hemorrhage patients, who were admitted to the Hiroshima University Hospital and the Suiseikai Kajikawa Hospital, Japan, from January 2013 to April 2016, were enrolled in this prospective study. The study protocols were approved by the ethics committee of the Hiroshima University Hospital (Epd-614-2) and the Suiseikai Kajikawa Hospital (2015–3) and performed according to the guidelines of the national government based on the Helsinki Declaration of 1964. Written informed consent was obtained from all patients or their relatives. All data analyses were conducted in a blinded manner.

We included patients who were admitted within 7 days from onset, were aged ≥ 20 years, and for whom consent to participate in this study was obtained from the patient or their relatives. We excluded patients who could not undergo head computed tomography (CT) and magnetic resonance imaging (MRI). We also excluded the patients who were diagnosed with hemorrhagic infarction or trauma-induced hemorrhages.

Data acquisition

Baseline clinical characteristic data, including age, sex, body mass index (BMI), comorbidities (hypertension, diabetes mellitus, dyslipidemia, atrial fibrillation, chronic kidney disease, and stroke), smoking and drinking habits, medication of antithrombotic drugs (anti-platelet and anticoagulant), onset to admission time, blood pressure, and C-reactive protein (CRP) levels were collected from all patients. Two stroke specialists (SA and EI) evaluated the stroke severity and conscious level. Stroke severity on admission was evaluated using the NIH Stroke Scale (NIHSS). Conscious level was evaluated with Glasgow coma scale. Comorbidities were defined according to a previous report [13] based on the Japanese hypertension, diabetes mellitus, dyslipidemia, atrial fibrillation, and chronic kidney disease guidelines. Hypertension was defined as the use of anti-hypertensive medication before admission or confirmed blood pressure of ≥140/90 mmHg at rest measured 2 weeks after onset. Diabetes mellitus was defined as a glycated hemoglobin level of ≥6.5%, fasting blood glucose level of ≥126 mg/dL, or use of anti-diabetes medication. Dyslipidemia was defined as a total cholesterol level of ≥220 mg/dL, low-density lipoprotein cholesterol level of ≥140 mg/dL, high-density lipoprotein cholesterol level of <40 mg/dL, triglyceride levels of ≥150 mg/dL, or use of anti-hyperlipidemia medication. Atrial fibrillation was defined as follows: (1) a history of sustained or paroxysmal atrial fibrillation or (2) atrial fibrillation detection on arrival or during admission. Renal functioning was calculated with the estimated glomerular filtration rate (eGFR) using a revised equation for the Japanese population as follows: eGFR (mL/min/1.73 m2) = 194 × serum creatinine−1.094 × age−0.287 × 0.739 (for women) [14]. Chronic kidney disease was defined as an eGFR <60 mL/min/1.73 m2. Imaging analysis with head computed tomography (CT) and magnetic resonance imaging (MRI) was performed in all patients for acute cerebral hemorrhage diagnosis. The cerebral hematoma was evaluated using CT. The admission and follow-up CT scans (24 hours after admission) were performed with axial 5-mm section thickness. Two experienced neurologists (MN and YS) measured the cerebral hematomas using the ABC/2 formula. The neuroimaging evaluation remained blinded from the clinical assessment. Based on the criteria used in several large clinical studies, hematoma growth was defined as an increase in hematoma volume of >33% or >12.5 mL at 24-hour follow-up [15]. The other two experienced neurologists (NK and KI) also performed evaluations of MRI findings and hematoma growth neuroimaging predictors, which were detected as the blend sign or black hole sign using plain head CT [15]. Cerebral amyloid angiopathy (CAA) was diagnosed using modified Boston criteria [16].

We collected clinical data regarding the acute phase, including intraventricular hemorrhage extension, pharmacological blood pressure management during the first 24 hours, surgical management approach, tube ventilatory use, and septic complications.

When we evaluated the 3-month outcome, we excluded patients who were disabled prior to stroke (corresponding to premorbid modified Rankin Scale [mRS] score ≥2). An unfavorable 3-month outcome was defined as a 3 or higher on the mRS.

Measurement of serum antibody titers of periodontal pathogens

Serum IgG antibody titers of periodontal pathogens were determined using ELISA as previously described [11]. Bacterial antigen-coated wells were washed with phosphate-buffered saline with Tween (PBST); serum samples in PBST were then added to the wells. After incubation at 4°C overnight, the wells were washed with PBST and filled with alkaline phosphatase-conjugated goat anti-human IgG (gamma-chain specific, Abcam, Cambridge, MO) in PBST. After another incubation at 37°C for 2 hours, the wells were again washed with PBST, an aliquot of p-nitrophenylphosphate at 1 mg/mL (WAKO Pure Chemical Industries Ltd., Osaka, Japan) in 10% diethanolamine buffer was added to each well as a substrate, and incubation was performed at 37°C for 30 minutes. Optical density at 405 nm was measured using a microplate reader (iMark, Bio-Rad Laboratories Inc., Hercules, CA). Serum samples were collected from patients within 3 days after stroke onset and stored at −80°C. Sonicated preparations of the following periodontal pathogens were used as bacterial antigens Porphyromonas (P.) gingivalis, Aggregatibacter (A.) actinomycetemcomitans, P. intermedia, Prevotella nigrescens, F. nucleatum, Treponema denticola, Tannerella forsythensis, Campylobacter rectus, and Eikenella corrodens. We selected these representative periodontal pathogens based on the previously reported association of serum antibody titers with stroke outcome [13]. The serum of 5 healthy individuals was pooled and used for calibration. Using serial dilutions of the pooled control serum, the standard reaction was defined based on the ELISA unit (EU) such that 100 EU corresponded to a 1:3200 dilution of the calibrator sample. For statistical analysis, we used the common logarithms of serum IgG antibody titers.

Statistical analysis

Data are expressed as the mean ± standard deviation or the median (minimum, maximum) for continuous variables, and frequencies and percentages for discrete variables. Statistical analysis was performed using the JMP 14.0 statistical software (SAS Institute Inc., Cary, NC, USA). The statistical significance of intergroup differences was assessed using the unpaired t-test or Mann-Whitney U test for continuous variables or the Fisher exact test or χ2 test for discrete variables as appropriate. Because there were no reports of hemorrhagic stroke, we calculated the sample size according to the past investigations for the IgG titers of periodontal pathogens in atherothrombotic stroke [9]. Based on an alpha level = 0.05 and power = 0.80, we estimated that we would require a total of n = 99 participants. Baseline data of cerebral hemorrhage patients were analyzed, and two-step strategies were employed to assess the relative importance of variables in their association with hemorrhage growth and poor outcome using least square linear regression analysis. We first performed a univariate analysis, followed by a multi-factorial analysis with selected factors with p < 0.05 in the former analysis. We considered p < 0.05 as statistically significant. We also calculated the statistical power and effect size as post hoc analysis.

Results

A total of 115 patients (44 females, aged 71.3 ± 13.1 years) with acute cerebral hemorrhage were registered in this study. The baseline clinical characteristics are shown in Table 1. Among them, 32 (27.8%) patients had histories of stroke, 16 of whom were cerebral hemorrhage. The number of patients with anti-platelet and anticoagulant drug use was 20 (17.4%) and 15 (13.2%), respectively. The median time from onset to admission was 147 minutes (min–max: 26–7200). CAA was diagnosed 16 (13.9%) patients, of whom 7 patients were probable CAA and 6 patients were possible CAA. The mean serum IgG titers of periodontal disease pathogens from all patients are summarized in Table 2.

Table 1

Baseline clinical characteristics.

	n = 115
Age, mean±SD	71.3±13.1
Sex [female], n (%)	44 (38.3)
Body mass index, kg/m2, mean±SD	22.3±4.1
History
Hypertension, n (%)	96 (83.5)
Diabetes mellitus, n (%)	21 (18.3)
Dyslipidemia, n (%)	30 (26.1)
Atrial fibrillation, n (%)	13 (11.3)
Chronic kidney disease, n (%)	35 (30.4)
Stroke, n (%)	32 (27.8)
cerebral hemorrhage, n (%)	16 (50.0)*
Duration from the past stroke, month, median (minimum, maximum)	36 (0.67, 384)*
Current smoker, n (%)	52 (47.3)
Habitual drinker, n (%)	30 (27.3)
Usage of anti-platelet drug, n (%)	20 (17.4)
Usage of anticoagulant drug, n (%)	15 (13.2)
Time from onset to admission, minutes, median (minimum, maximum)	147 (26, 7200)
Systolic blood pressure on admission, mmHg, mean±SD	175.8±30.8
Diastolic blood pressure on admission, mmHg, mean±SD	100.0±17.9
NIHSS score, median (minimum, maximum)	9 (0, 38)
Glasgow coma scale, median (minimum, maximum)	15 (4, 15)
CRP, mg/dl, mean±SD	0.75±1.31
Premorbid mRS ≥2, n (%)	22 (19.1)
Cerebral hematoma on admission
Volume, ml, mean±SD	15.5±20.6
Supratentorial, n (%)	94 (81.7)
intraventricular hemorrhage extension, n (%)	38 (33.0)
hematoma growth neuroimaging predictors, n (%)	13 (11.3)
Pharmacological blood pressure management during the first 24 hours, n (%)	92 (80.0)
Surgical Management approach, n (%)	9 (7.8)
Tube ventilatory use, n (%)	9 (7.8)
In hospital septic complications, n (%)	14 (12.2)
Etiology
Hypertensive, n (%)	92 (80.0)
Cerebral amyloid angiopathy, n (%)	16 (13.9)
Others, n (%)	7 (6.1)

SD, standard deviation; NIHSS, NIH Stroke Scale; CRP, C-reactive protein; mRS, modified Rankin Scale.

*n = 32.

Table 2

Mean serum IgG titers of periodontal disease pathogens in all cerebral hemorrhage patients.

	IgG titer (EU)
Porphyromonas gingivalis	656.0 ± 1949.9
Aggregatibacter actinomycetemcomitans	185.1 ± 454.2
Prevotella intermedia	131.6 ± 191.1
Prevotella nigrescens	74.3 ± 111.0
Fusobacterium nucleatum	54.8 ± 87.1
Treponema denticola	43.0 ± 88.0
Tannerella forsythensis	41.3 ± 88.7
Campylobacter rectus	201.2 ± 511.3
Eikenella corrodens	61.1 ± 92.4

EU, ELISA unit.

We found hemorrhage growth in 13 patients (11.3%). The potential factors associated with hemorrhage growth were evaluated using the univariate analysis (listed in Tables 1 and 2). In this analysis, hemorrhage growth was associated with the history of atrial fibrillation, usage of anticoagulant, and the IgG titers of A. actinomycetemcomitans. Among patients with hemorrhage growth, 5 (38.5%) patients had used an anticoagulant, namely warfarin, and were not injected with the prothrombin complex concentrate during the time of admission due to the Japanese medical insurance system at that time. Multivariate logistic regression analysis revealed that usage of anticoagulant (odds ratio 8.36, 95% CI 1.35–51.70, p = 0.02), septic complications (odds ratio 10.20, 95% CI 1.94–53.72, p = 0.01), and the IgG titer of A. actinomycetemcomitans (odds ratio 5.26, 95% CI 1.52–18.25, p = 0.01) were independently associated with hemorrhage growth (Table 3). Regarding the IgG titer of A. actinomycetemcomitans, the statistical power and effect size of Cohen’s d were 0.80 and 0.67, respectively.

Table 3

Factors influencing cerebral hemorrhage growth.

	Hematoma growth		Univariate analysis	Multivariate analysis
	(+) n = 13	(-) n = 102	p value	odds ratio	95% CI	p value
Age, mean±SD	74.0±7.8	71.0±13.6	0.43
Sex [female], n (%)	3 (23.1)	41 (40.2)	0.23
Body mass index, kg/m2, mean±SD	23.3±5.3	22.2±3.9	0.37
Hypertension, n (%)	11 (84.6)	85 (83.3)	0.91
Diabetes mellitus, n (%)	3 (23.1)	18 (17.7)	0.63
Dyslipidemia, n (%)	4 (30.8)	26 (26.5)	0.68
Atrial fibrillation, n (%)	4 (30.8)	9 (8.8)	0.02*	1.69	0.28–10.34	0.57
Chronic kidney disease, n (%)	5 (38.5)	30 (29.4)	0.50
Stroke, n (%)	6 (46.2)	26 (25.5)	0.12
Duration from the past stroke, n (%)	22 (12, 384)	60 (0.67, 312)	0.33
Current smoker, n (%)	5 (38.5)	47 (48.5)	0.50
Habitual drinker, n (%)	3 (23.1)	27 (27.8)	0.72
Usage of anti-platelet, n (%)	1 (7.7)	19 (18.6)	0.33
Usage of anticoagulant, n (%)	5 (38.5)	10 (9.9)	0.004*	8.36	1.35–51.70	0.02*
Time from onset to admission, minute, median (minimum, maximum)	240 (69, 5760)	135.5 (26, 7200)	0.11
Systolic blood pressure on admission, mmHg, mean±SD	175.9±22.3	175.8±31.8	0.99
Diastolic blood pressure on admission, mmHg, mean±SD	99.8±13.3	100.1±18.4	0.97
NIHSS score, median (minimum, maximum)	16 (0, 38)	9 (0, 38)	0.17
Glasgow coma scale, median (minimum, maximum)	14 (4, 15)	15 (4, 15)	0.33
CRP, mg/dl, mean±SD	1.30±1.94	0.68±1.21	0.11
Cerebral hematoma volume on admission, ml, mean±SD	9.47±8.26	16.29±21.53	0.26
Supratentorial hematoma, n (%)	10 (76.9)	84 (82.4)	0.63
Intraventricular hemorrhage extension, n (%)	7 (53.9)	31 (30.4)	0.09
Hematoma growth neuroimaging predictors, n (%)	1 (7.7)	12 (11.8)	0.66
Pharmacological blood pressure management during the first 24 hours, n (%)	13 (100.0)	79 (77.5)	0.06
Surgical Management approach, n (%)	1 (7.7)	8 (7.8)	0.98
Tube ventilatory use, n (%)	2 (15.4)	7 (6.9)	0.28
In hospital septic complications, n (%)	4 (30.8)	10 (9.8)	0.03*	10.2	1.94–53.72	0.01*
Hypertensive cerebral hemorrhage, n (%)	12 (92.3)	80 (78.4)	0.24
Cerebral amyloid angiopathy, n (%)	0 (0)	16 (15.7)	0.12
Other etiologies, n (%)	1 (7.7)	6 (5.9)	0.80
IgG titer of periodontal disease pathogen
log P. gingivalis, mean±SD	2.10±1.03	2.01±0.91	0.75
log A. actinomycetemcomitans, mean±SD	2.15±0.78	1.69±0.67	0.02*	5.26	1.52–18.25	0.01*
log P. intermedia, mean±SD	2.07±0.73	1.76±0.58	0.08
log P. nigrescens, mean±SD	1.48±0.87	1.45±0.73	0.91
log F. nucleatum, mean±SD	1.62±0.61	1.39±0.54	0.21
log T. denticola, mean±SD	1.30±0.46	1.10±0.75	0.34
log T. forsythensis, mean±SD	1.38±0.52	1.23±0.60	0.38
log C. rectus, mean±SD	1.92±0.75	1.71±0.70	0.31
log E. corrodens, mean±SD	1.73±0.50	1.45±0.59	0.11

CI, confidence interval; SD, standard deviation; NIHSS, NIH Stroke Scale; CRP, C-reactive protein.

* means p < 0.05.

When we evaluated the 3-month outcome, 22 patients were excluded based on premorbid mRS scores of ≥2. The potential factors associated with poor outcome (mRS score ≥3) (listed in Tables 1–3) were evaluated using the univariate analysis. We found that poor outcome was associated with age, NIHSS score, cerebral hematoma volume, cerebral hematoma growth, and IgG titers of F. nucleatum. Multivariate logistic regression analysis revealed that age (odds ratio 1.09, 95% CI 1.01–1.17, p = 0.02), NIHSS score (odds ratio 1.29, 95% CI 1.09–1.52, p = 0.002), and the IgG titer of F. nucleatum (odds ratio 7.86, 95% CI 1.08–57.08, p = 0.04) were independently associated with poor outcome, but not cerebral hematoma volume or growth (Table 4). Regarding the IgG titer of F. nucleatum, the statistical power and effect size of Cohen’s d were 0.91 and 0.69, respectively.

Table 4

Factors influencing unfavorable outcome (modified Rankin Scale ≥3).

	Outcome		Univariate analysis	Multivariate analysis
	Favorable, n = 48	Unfavorable, n = 45	p value	odds ratio	95% CI	p value
Age, mean±SD	66.9±12.1	73.4±11.5	0.01*	1.09	1.01–1.17	0.02*
Sex(female), n (%)	14 (29.2)	19 (42.2)	0.23
Body mass index, kg/m2, mean±SD	22.8±4.0	22.6±4.4	0.83
Hypertension, n (%)	42 (87.5)	37 (82.2)	0.48
Diabetes mellitus, n (%)	7 (14.6)	10 (22.2)	0.34
Dyslipidemia, n (%)	16 (30.1)	12 (26.7)	0.48
Atrial fibrillation, n (%)	5 (10.4)	5 (11.1)	0.91
Stroke, n (%)	8 (16.7)	10 (22.2)	0.50
Duration from the past stroke, month, median (minimum, maximum)	54 (12, 312)	60 (0.8, 384)	0.70
Current smoker, n (%)	28 (59.6)	17 (41.5)	0.09
Habitual drinker, n (%)	16 (34.0)	10 (24.4)	0.32
Usage of anti-platelet drug, n (%)	8 (16.7)	6 (13.3)	0.65
Usage of anticoagulant, n (%)	4 (8.3)	7 (15.6)	0.28
Time from onset to admission, minute, median (minimum, maximum)	143.5 (26, 7200)	153 (29, 5760)	0.95
Systolic blood pressure on admission, mmHg, mean±SD	177.3±31.0	175.3±30.2	0.75
Diastolic blood pressure on admission, mmHg, mean±SD	100.7±18.5	100.5±17.4	0.95
NIHSS score, median (minimum, maximum)	3 (0, 19)	14 (0, 29)	<0.001*	1.29	1.09–1.52	0.002*
Glasgow coma scale, median (minimum, maximum)	15 (11, 15)	14 (4, 15)	<0.001*	1.57	0.98–2.53	0.06
CRP, mg/dl, mean±SD	0.50±1.00	0.97±1.63	0.09
Cerebral hematoma volume on admission, ml, mean±SD	6.63±9.36	19.75±29.21	<0.001*	1.00	0.93–1.08	0.90
Cerebral hematoma growth, n (%)	1 (2.1)	9 (20.0)	0.01*	14.08	0.98–202.21	0.05
Supratentorial hematoma, n (%)	36 (75.0)	38 (84.4)	0.26
Intraventricular hemorrhage extension, n (%)	5 (10.4)	21 (46.7)	<0.001*	5.17	0.93–28.67	0.06
hematoma growth neuroimaging predictors, n (%)	2 (4.2)	5 (11.1)	0.20
Pharmacological blood pressure management during the first 24 hours, n (%)	34 (70.8)	39 (86.7)	0.06
Surgical Management approach, n (%)	1 (2.1)	5 (11.1)	0.08
Tube ventilatory use, n (%)	1 (2.1)	5 (11.1)	0.08
In hospital septic complications, n (%)	1 (2.1)	8 (17.8)	0.01*	8.72	0.47–162.50	0.15
Hypertensive cerebral hemorrhage, n (%)	37 (77.1)	38 (84.4)	0.37
Cerebral amyloid angiopathy, n (%)	8 (16.7)	4 (8.9)	0.26
Other etiologies, n (%)	3 (6.3)	3 (6.7)	0.93
IgG titer of periodontal disease pathogen
log P. gingivalis, mean±SD	1.86±1.01	2.20±0.74	0.07
log A. actinomycetemcomitans, mean±SD	1.78±0.81	1.73±0.64	0.73
log P. intermedia, mean±SD	1.69±0.68	1.86±0.55	0.19
log P. nigrescens, mean±SD	1.44±0.82	1.50±0.72	0.74
log F. nucleatum, mean±SD	1.28±0.48	1.63±0.51	<0.001*	7.86	1.08–57.08	0.04*
log T. denticola, mean±SD	1.11±0.75	1.21±0.67	0.47
log T. forsythensis, mean±SD	1.29±0.54	1.25±0.62	0.74
log C. rectus, mean±SD	1.70±0.67	1.78±0.74	0.57
log E. corrodens, mean±SD	1.42±0.63	1.49±0.58	0.59

An unfavorable 3-month outcome was defined as a 3 or higher on the modified Rankin Scale.

CI, confidence interval; SD, standard deviation; NIHSS, NIH Stroke Scale; CRP, C-reactive protein.

* means p < 0.05.

Discussion

Our findings suggest that periodontal disease may be associated with cerebral hemorrhage and its clinical course. Specifically, we reveal for the first time that serum IgG titers of a few particular periodontal pathogens may be useful biomarkers for predicting the clinical course of a cerebral hemorrhage.

In the present study, we provide evidence that an elevated serum IgG titer of A. actinomycetemcomitans is an independent factor for predicting hemorrhage growth. A. actinomycetemcomitans is a gram-negative, facultatively anaerobic coccobacillus and is considered the major etiologic agent of localized aggressive periodontitis [17]. It also contributes to chronic periodontitis, and according to previous reports, an elevated serum titer of A. actinomycetemcomitans predicts stroke and coronary heart disease [18-20]. A potential explanation for these relationships is that individuals infected with A. actinomycetemcomitans harbor T-cells specific to this bacterium in their blood, which express the receptor activator of nuclear factor-κB (RANK) ligand [21]. This ligand stimulates the vascular smooth muscle cells to produce matrix metalloproteinase-9, which may promote the growth of cerebral hemorrhages.

We further demonstrate that the serum IgG titer of F. nucleatum independently predicted an unfavorable outcome in cerebral hemorrhage, which is consistent with a previous study [13]. Fusobacterium nucleatum is strictly an anaerobic gram-negative rod bacterium, normally found in the oral cavity. It is considered to be a periodontal pathogen because it is frequently isolated from periodontitis lesions, produces a high number of tissue irritants, and often aggregates with other periodontal pathogens as a bridge between early and late colonizers [22,23]. Fusobacterium nucleatum, which can pass the blood-brain barrier, is associated with colon cancer and Alzheimer’s disease [3,12], and was found to cause brain abscesses [24,25]. This bacterium also has abilities to adhere to and invade host vascular endothelial cells using the protein, Fusobacterium adhesin A (FadA), which binds to vascular endothelial-cadherin on the cell surface, thereby triggering a breakdown of endothelial cell-to-cell junctions [26]. The virulence of F. nucleatum mediates endotoxin activity, hemagglutination, as well as aggregation and death of immune cells via their outer membrane proteins, such as fibroblast activation protein 2 (FAP-2) and radiation genes (RadD) [27]. In addition, F. nucleatum causes an increase in the expression of genes associated with host immune responses, such as inflammatory chemokines and cytokines. Together, these reports suggest that F. nucleatum may have a systemic adverse impact on stroke patients apart from hemorrhage growth.

There were some limitations to this study. First, there was the issue of sample size and sampling bias. In this study, the sample size was modest. Thus, despite the multicenter study design, sample size calculation, and post hoc analysis, some sampling bias might exist. The average age of patients and the numbers of patients with CAA, previous stroke, and taking antithrombotic drugs were also relatively high, whereas the total frequency of intracerebral hematoma growth was low. Regarding this low intracerebral hematoma growth frequency, the time from onset to admission might have affected the results. In this study, we included patients within 7 days from onset. However, all patients were considered during the acute phase, and the median time from onset to admission was 147 minutes. Time from onset to collection of blood sample also varied because of the study design. However, all patients were tested within a week from onset. Since periodontal disease is a chronic disease, mild variation might not affect the results. Second, this was a cross-sectional observational study, which makes it difficult to adequately assess the biological relationship between periodontal pathogens and cerebral hemorrhage. To provide more conclusive evidence, in vivo animal experiments are required. If periodontal pathogens themselves are risk factors for a malignant outcome of cerebral hemorrhage, daily oral care and regular dental examination could practically improve the clinical course. Third, serum IgG titers were used to investigate the association between periodontal disease and the clinical course of a cerebral hemorrhage, but the severity of periodontal disease and the intraoral environment was not directly evaluated. Therefore, future studies could consider these parameters and assess how they affect cerebral hemorrhage pathology and clinical outcomes. While the influence of serum IgG titers of periodontal pathogen on cerebral hemorrhage has not been fully determined, specific periodontal pathogen infection can be a useful biomarker for predicting the clinical course of cerebral hemorrhage.

Stroke mortality rates have been decreasing owing to the development of advanced medical treatments; however, disability rates of stroke survivors have been increasing. Given our findings of an association between periodontal pathogen titers and the clinical course of a cerebral hemorrhage, it is important to seriously consider each of these pathogens and their potential negative impact on affected individuals.

Conclusions

Our observations reveal the impact of two periodontal pathogens on cerebral hemorrhage, namely that elevated serum IgG titers of A. actinomycetemcomitans predicted hemorrhage growth and that those of F. nucleatum predicted a poor outcome in patients with this disease. Consequently, periodontal pathogens could play an important role in the management and treatment of stroke patients, for which the assessment of IgG titers has shown to be a valid tool.

COI.

(PDF)

Click here for additional data file.

All relevant data of the study.

(XLSX)

Click here for additional data file.

3 Jun 2020

PONE-D-20-12711

Serum IgG titers of periodontal pathogens predict cerebral hemorrhage growth and 3-month outcome

PLOS ONE

Dear Dr. Hosomi,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

This is a very interesting manuscript, addressing an uncommon but reasonable association among pathogens and their possible role in hemorrhagic stroke behavior, but many major issues arising from methodology and results, should be analyzed before considering this manuscript as suitable for publication.

Please submit your revised manuscript by Jul 18 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Miguel A. Barboza, MD, MSc

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following in the Competing Interests section:

"Hirofumi Maruyama reports research support from Eisai, Pfizer, Takeda

Pharmaceutical, Otsuka Pharmaceutical, Nihon Pharmaceutical, Shionogi, Teijin

Pharma, Fuji Film, Boehringer Ingelheim, Sumitomo Dainippon Pharma, Nihon Medi-

Physics, Bayer, MSD, Daiichi Sankyo, Kyowa Hakko Kirin, Sanofi, Novartis, Kowa

Pharmaceutical, Astellas Pharma, Tsumura, Japan Blood Products Organization,

Mitsubishi Tanabe Pharma, Mylan, which are unrelated to the submitted work. All other

authors declare that they have no conflicts of interest."

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors ).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests:

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: With glad and interest I have reed the manuscript titled. .Serum IgG titers of periodontal pathogens predict cerebral hemorrhage growth and 3- month outcome

The study is relevant for focusing on five pivotal aspects:

a.- The study on intracerebral hemorrhage (ICH)

b.- The study of periodontal disease

c.- Hematoma growth or hematoma enlargement.

d.- The relationship between them

e.- The short-term outcome

In a prospective (observational cross-sectional) design study from two Japanese hospitals were authors evaluated the growth of hematoma and the short-term outcome in 115 patients with acute ICH and its possible association with the serum IgG titers of 9 of the most frequent periodontal pathogens.

In short, one periodontal pathogen was associated to intracerebraral growth hematoma and anotherone to 3-months poor outcome (A. actinomycetemcomitans and F. nucleatum respectively).

Methods.

Measurement of serum antibody titers of periodontal pathogens serum samples were collected from patients within 3 days after stroke onset.

Q.- Why 3 days after ICH onset?

Results

1.- Low modest sample

2.- Old age average of patients

3.- Low total frequency of intracerebraral growth hematoma (11.3%)

4.- Previos stroke and previos antithrombotic drus in 27.3% and 30.6% respectively

5.- Some patients with 5 days admission after ICH onset symptoms (maximum=7200 minutes) Table 1.

6.- Cerebral amyloid angiopathy 13.9% (definitive, probable or possible?)

Q_ Please, clarify and discuss this findings, especially those that may affect or bias the analysis of the results (hematoma growth and/or outcome)

Also:

1.- If previous stroke (please clarify type: ischemic or hemorrhagic)

2.- Time from previous stroke (please clarify: years or recently in days or months?)

3.- None patients with amyloid angiopathy observed hematoma growth. Please discuss this intrigant finding.

4.- Intraventricular ICH frequency, please clarify

5.- Surgical Management approach, please clarify frequency

6.- Frequency of tube ventilatory use or other orofaringeal devices

7.- In hospital septic complications frequency

Reviewer #2: This article addresses the association between serum IgG titers of periodontal pathogens as predictors for cerebral hemorrhage growth and poor functional outcome at 3 months (i.e. modified Ranking Scale �3). The topic is interesting and it has not been explored before on a prospective cross-sectional study for cerebral hemorrhage. The abstract is appropriate for the content of the text. The authors mentioned relevant information regarding the ‘non-availability’ for reversal agents for oral anticoagulants by the time the study underwent, and their awareness for lack of periodontal disease clinical status from participants. Although the analysis included the most common vascular risk factors, there are some clinical characteristics that should have been considered to adjust during the statistical analysis, such as: Glasgow Coma Scale score, intraventricular hemorrhage extension, and early hematoma growth neuroimaging predictors (e.g. blend sign). Furthermore, since early blood pressure control in acute intracerebral hemorrhage is a key prognostic factor, it would have been useful to be mentioned if any of the participants required pharmacological blood pressure management during the first 24 hours, and consider to adjust for this on the statistical analysis. The conclusions that “elevated serum IgG titers of Aggregatibacter actinomycetemcomitans is associated to cerebral hemorrhage growth at 24 hours follow-up from admission and yet no effect on functional outcome at 3-months follow-up, and that elevated IgG titers of Fusobacterium nucleatum predicted a poor outcome”, are reasonable given the data reported in the article, but overreached the collected data.

Please find attached my additional comments for the authors.

Best wishes.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Serum IgG periodontal pathogens and Cerebral Haemorrhage _ Reviewer .docx

Click here for additional data file.

26 Aug 2020

Thank you again for reviewing our manuscript. We appreciate the insightful comments and advice of the reviewers. We have provided point-by-point responses to each of the comments below and highlighted the corresponding revisions in the manuscript document in red.

Response to the editor

Comment 1: Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Response 1: We will ensure and meet the PLOS ONE's style requirements.

Comment 2: Thank you for stating the following in the Competing Interests section:

"Hirofumi Maruyama reports research support from Eisai, Pfizer, Takeda

Pharmaceutical, Otsuka Pharmaceutical, Nihon Pharmaceutical, Shionogi, Teijin

Pharma, Fuji Film, Boehringer Ingelheim, Sumitomo Dainippon Pharma, Nihon Medi-

Physics, Bayer, MSD, Daiichi Sankyo, Kyowa Hakko Kirin, Sanofi, Novartis, Kowa

Pharmaceutical, Astellas Pharma, Tsumura, Japan Blood Products Organization,

Mitsubishi Tanabe Pharma, Mylan, which are unrelated to the submitted work. All other

authors declare that they have no conflicts of interest."

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared. Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf. Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

Response 2: We added the following sentence in the Competing Interests section and included the Competing Interests statement in the cover letter:

This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Response to the reviewers

Reviewer #1: With glad and interest I have read the manuscript titled. Serum IgG titers of periodontal pathogens predict cerebral hemorrhage growth and 3- month outcome. The study is relevant for focusing on five pivotal aspects:

a.- The study on intracerebral hemorrhage (ICH)

b.- The study of periodontal disease

c.- Hematoma growth or hematoma enlargement.

d.- The relationship between them

e.- The short-term outcome

In a prospective (observational cross-sectional) design study from two Japanese hospitals were authors evaluated the growth of hematoma and the short-term outcome in 115 patients with acute ICH and its possible association with the serum IgG titers of 9 of the most frequent periodontal pathogens.

In short, one periodontal pathogen was associated to intracerebraral growth hematoma and anotherone to 3-months poor outcome (A. actinomycetemcomitans and F. nucleatum respectively).

Response: We appreciate your review and suggestions. To prevent as much bias as possible, we have added the clinical information which the reviewers suggested and analyzed it. In addition, we added the sample size calculation, statistical power, significance level, and effect size for statistical confirmation. However, there were some sampling biases, which we have mentioned as one of the limitations of the study.

Comment 1:

Methods.

Measurement of serum antibody titers of periodontal pathogens serum samples were collected from patients within 3 days after stroke onset.

Q.- Why 3 days after ICH onset?

Response 1: We tried to collect the serum samples as soon as possible. However, it took time to obtain the consent from the patients or their relatives. In this way, we decided on a time within 3 days due to the procedural issue. All patients were tested within a week from onset. Since periodontal disease is a chronic disease, mild variations in the sampling time might not affect the results. We added the following text to the ‘Discussion’ section:

Page 26, Lines 288-291

Time from onset to collection of blood sample also varied because of the study design. However, all patients were tested within a week from onset. Since periodontal disease is a chronic disease, mild variation might not affect the results.

Comment 2:

Results

1.- Low modest sample

2.- Old age average of patients

3.- Low total frequency of intracerebral growth hematoma (11.3%)

4.- Previous stroke and previous antithrombotic drugs in 27.3% and 30.6% respectively

5.- Some patients with 5 days admission after ICH onset symptoms (maximum=7200 minutes) Table 1.

6.- Cerebral amyloid angiopathy 13.9% (definitive, probable or possible?)

Q_ Please, clarify and discuss this findings, especially those that may affect or bias the analysis of the results (hematoma growth and/or outcome)

Response 2: We appreciate your suggestions. In this study, the sample size was modest. Thus, despite the multicenter study design, sample size calculation, and post hoc analysis, some sampling bias might exist. As the reviewer mentioned, the average age of patients was high. The numbers of patients with CAA (7 patients were probable CAA and 6 patients were possible CAA), previous stroke, and taking antithrombotic drugs were also relatively high, whereas the total frequency of intracerebral hematoma growth was low. Regarding this low intracerebral hematoma growth frequency, the time from onset to admission might have affected the results. In this study, we included patients within 7 days from onset. As the reviewer pointed, one patient was admitted 5 days after intracerebral hemorrhage onset. Especially for patients with CAA, admission after onset tended to be delayed, because the symptoms were mild regardless of the bleeding. We understand that time from onset to admission affects hematoma growth. However, all patients were considered during the acute phase, and the median time from onset to admission was 137 minutes. To address these issues, we have modified the Materials and Methods and Results sections and discussed the possible sampling biases as limitations in the Discussion. We also added the relevant clinical information in Tables 1, 3, and 4.

Page 8, Lines 107-110

Conscious level was evaluated with Glasgow coma scale. Comorbidities were defined according to a previous report [13] based on the Japanese hypertension, diabetes mellitus, dyslipidemia, atrial fibrillation, and chronic kidney disease guidelines.

Page 9, Lines 130-137

The other two experienced neurologists (NK and KI) also performed evaluations of MRI findings and hematoma growth neuroimaging predictors, which were detected as the blend sign and black hole sign using plain head CT [15]. Cerebral amyloid angiopathy (CAA) was diagnosed using modified Boston criteria [16].

We collected clinical data regarding the acute phase, including intraventricular hemorrhage extension, pharmacological blood pressure management during the first 24 hours, surgical management approach, tube ventilatory use, and septic complications.

Page 12, Lines 173-176

Because there were no reports of hemorrhagic stroke, we calculated the sample size according to the past investigations for the IgG titers of periodontal pathogens in atherothrombotic stroke [9]. Based on an alpha level = 0.05 and power = 0.80, we estimated that we would require a total of n = 99 participants.

Page 12, Lines 181-182

We considered p < 0.05 as statistically significant. We also calculated the statistical power and effect size as post hoc analysis.

Pages 12-13, Lines 187-191

Among them, 32 (27.8%) patients had histories of stroke, 16 of whom were cerebral hemorrhage. The number of patients with anti-platelet and anticoagulant drug use was 20 (17.4%) and 15 (13.2%), respectively. The median time from onset to admission was 147 minutes (min–max: 26–7200). CAA was diagnosed 16 (13.9%) patients, of whom 7 patients were probable CAA and 6 patients were possible CAA.

Pages 15-16, Lines 210-216

Multivariate logistic regression analysis revealed that usage of anticoagulant (odds ratio 8.36, 95% CI 1.35–51.70, p = 0.02), septic complications (odds ratio 10.20, 95% CI 1.94–53.72, p = 0.01), and the IgG titer of A. actinomycetemcomitans (odds ratio 5.26, 95% CI 1.52–18.25, p = 0.01) were independently associated with hemorrhage growth (Table 3). Regarding the IgG titer of A. actinomycetemcomitans, the statistical power and effect size of Cohen’s d were 0.80 and 0.67, respectively.

Page 20, Lines 227-233

Multivariate logistic regression analysis revealed that age (odds ratio 1.09, 95% CI 1.01–1.17, p = 0.02), NIHSS score (odds ratio 1.29, 95% CI 1.09–1.52, p = 0.002), and the IgG titer of F. nucleatum (odds ratio 7.86, 95% CI 1.08–57.08, p = 0.04) were independently associated with poor outcome, but not cerebral hematoma volume or growth (Table 4). Regarding the IgG titer of F. nucleatum, the statistical power and effect size of Cohen’s d were 0.91 and 0.69, respectively.

Page 26, Lines 279-291

There were some limitations to this study. First, there was the issue of sample size and sampling bias. In this study, the sample size was modest. Thus, despite the multicenter study design, sample size calculation, and post hoc analysis, some sampling bias might exist. The average age of patients and the numbers of patients with CAA, previous stroke, and taking antithrombotic drugs were also relatively high, whereas the total frequency of intracerebral hematoma growth was low. Regarding this low intracerebral hematoma growth frequency, the time from onset to admission might have affected the results. In this study, we included patients within 7 days from onset. However, all patients were considered during the acute phase, and the median time from onset to admission was 147 minutes. Time from onset to collection of blood sample also varied because of the study design. However, all patients were tested within a week from onset. Since periodontal disease is a chronic disease, mild variation might not affect the results.

Comment 3:

Also:

1.- If previous stroke (please clarify type: ischemic or hemorrhagic)

2.- Time from previous stroke (please clarify: years or recently in days or months?)

3.- None patients with amyloid angiopathy observed hematoma growth. Please discuss this intrigant finding.

4.- Intraventricular ICH frequency, please clarify

5.- Surgical Management approach, please clarify frequency

6.- Frequency of tube ventilatory use or other orofaringeal devices

7.- In hospital septic complications frequency

Response 3: We appreciate your suggestions. To prevent as much bias as possible, we have added the clinical information which the reviewers suggested and analyzed it. We added the information of previous stroke (ischemic or hemorrhagic), time from previous stroke, intraventricular ICH frequency, surgical management approach, frequency of tube ventilatory use, and septic complications frequency. Indeed, it was intriguing that none of the patients with amyloid angiopathy had any hematoma growth. One possible reason for this is that in the patients with CAA, admission after onset tended to be delayed, because the symptoms were mild regardless of the bleeding. Sampling bias might also affect the results. Thus, we described the sampling biases as limitations in the ‘Discussion’ section (the changes were described as the previous response). We have also revised Tables 1, 3, and 4 accordingly.

Reviewer #2:

This article addresses the association between serum IgG titers of periodontal pathogens as predictors for cerebral hemorrhage growth and poor functional outcome at 3 months (i.e. modified Ranking Scale �3). The topic is interesting and it has not been explored before on a prospective cross-sectional study for cerebral hemorrhage. The abstract is appropriate for the content of the text. The authors mentioned relevant information regarding the ‘non-availability’ for reversal agents for oral anticoagulants by the time the study underwent, and their awareness for lack of periodontal disease clinical status from participants. Although the analysis included the most common vascular risk factors, there are some clinical characteristics that should have been considered to adjust during the statistical analysis, such as: Glasgow Coma Scale score, intraventricular hemorrhage extension, and early hematoma growth neuroimaging predictors (e.g. blend sign). Furthermore, since early blood pressure control in acute intracerebral hemorrhage is a key prognostic factor, it would have been useful to be mentioned if any of the participants required pharmacological blood pressure management during the first 24 hours, and consider to adjust for this on the statistical analysis. The conclusions that “elevated serum IgG titers of Aggregatibacter actinomycetemcomitans is associated to cerebral hemorrhage growth at 24 hours follow-up from admission and yet no effect on functional outcome at 3-months follow-up, and that elevated IgG titers of Fusobacterium nucleatum predicted a poor outcome”, are reasonable given the data reported in the article, but overreached the collected data.

Below there are more specific comments that should be clarified/addressed:

Response: We appreciate for your review and suggestions. To prevent the bias as possible, we included the clinical information such as Glasgow Coma Scale score, intraventricular hemorrhage extension, early hematoma growth neuroimaging predictors, if any of the pharmacological blood pressure management during the first 24 hours, time from previous stroke, surgical management approach, tube ventilatory use, and septic complication. In addition, we added the sample size calculation, statistical power, significance level and effect size for statistical confirmation.

Major issues:

Comment 1: P. 10: Please provide prevalence estimates regarding the IgG titers of periodontal pathogens in hemorrhagic stroke taken from the literature. More information needs to be added about the sample size calculation, statistical power, significance level and effect size. The sample size needs to be more extensively explained by the authors; this will allow to be precise about sizes effect (if any) in order to make solid conclusions and give opportunity to generate new hypothesis for future research.

Response 1: Because there were no reports regarding to the hemorrhagic stroke, we calculated the sample size according to the past investigations for IgG titers of periodontal pathogens in atherothrombotic stroke. We have added the statistical power, significance level, and effect size to the revised the manuscript and added statements regarding the IgG titers as follows:

Page 12, Lines 173-182

Because there were no reports of hemorrhagic stroke, we calculated the sample size according to the past investigations for the IgG titers of periodontal pathogens in atherothrombotic stroke [9]. Based on an alpha level = 0.05 and power = 0.80, we estimated that we would require a total of n = 99 participants. Baseline data of cerebral hemorrhage patients were analyzed, and two-step strategies were employed to assess the relative importance of variables in their association with hemorrhage growth and poor outcome using least square linear regression analysis. We first performed a univariate analysis, followed by a multi-factorial analysis with selected factors with p < 0.05 in the former analysis. We considered p < 0.05 as statistically significant. We also calculated the statistical power and effect size as post hoc analysis.

Pages 15-16, Lines 210-216

Multivariate logistic regression analysis revealed that usage of anticoagulant (odds ratio 8.36, 95% CI 1.35–51.70, p = 0.02), septic complications (odds ratio 10.20, 95% CI 1.94–53.72, p = 0.01), and the IgG titer of A. actinomycetemcomitans (odds ratio 5.26, 95% CI 1.52–18.25, p = 0.01) were independently associated with hemorrhage growth (Table 3). Regarding the IgG titer of A. actinomycetemcomitans, the statistical power and effect size of Cohen’s d were 0.80 and 0.67, respectively.

Page 20, Lines 227-233

Multivariate logistic regression analysis revealed that age (odds ratio 1.09, 95% CI 1.01–1.17, p = 0.02), NIHSS score (odds ratio 1.29, 95% CI 1.09–1.52, p = 0.002), and the IgG titer of F. nucleatum (odds ratio 7.86, 95% CI 1.08–57.08, p = 0.04) were independently associated with poor outcome, but not cerebral hematoma volume or growth (Table 4). Regarding the IgG titer of F. nucleatum, the statistical power and effect size of Cohen’s d were 0.91 and 0.69, respectively.

Comment 2: P. 12: Table 2. Select a more concise heading. It is not clear if the serum IgG titers makes reference to the ‘reference range’ from the five healthy volunteers or if these measurements are considered significant positive titers. It could be helpful to add a brief explanation on the ‘Results’ section.

Response 2: In Table 2, we show the mean ± SD of the serum IgG titers from all subjects in the study. As mentioned in the ‘Materials and Methods’ section, the standard reaction was defined based on the ELISA unit (EU) such that 100 EU corresponded to a 1:3200 dilution of the calibrator sample (5 healthy controls). Because there are no defined positive cutoff values, we reported the mean ± SD of the serum IgG titers from all subjects in the study. We have rewritten the heading and part of the ‘Results’ section.

Page 13, Lines 192-193.

The mean serum IgG titers of periodontal disease pathogens from all patients are summarized in Table 2.

Comment 3: P. 14: Table 3. It would be more informative if you add to the ‘P values’ the odds ratio and confidence intervals for the univariate and multivariable analysis. Also, notice that data from the table considered four patients with history of atrial fibrillation and five patients taking anticoagulants on the positive hematoma growth group. On this same group, from 13 patients with positive hematoma growth, twelve had a hypertensive etiology, zero amyloid angiopathy and the remaining one participant is not mentioned on the table data. It is relevant that authors address this on the results section and verified data in this table.

Response 3: We have added the odds ratio and confidence intervals into Table 3. One patient without atrial fibrillation took anticoagulants because of risk of cardioembolism due to left ventricular dysfunction. There were some cerebral hemorrhage patients with other etiology such as arteriovenous malformation and moyamoya disease. We also added a row for other etiologies in Table 3.

Comment 4: P. 17: Table 4. It would be more informative if you add to the ‘P values’ the odds ratio and confidence intervals for the univariate and multivariable analysis.

Response 4: In accordance with your suggestion, we have incorporated the odds ratio and confidence intervals into Table 4.

Minor issues:

1. Introduction:

Comment 5: P. 5: Please add more background information about the potential mechanism proposed for the influence of periodontal disease in stroke since this will give more context to the reader. It is required more detail about the large cohort study that demonstrated the association between periodontal disease and the incidence of specific etiologies for stroke (i.e. include hazard ratio and confidence interval for these statements).

Response 5: We appreciate your suggestion. We have added more background information about the influence of periodontal disease in stroke as follows:

Pages 5-6, Lines 58-77

Meta-analysis from previous studies showed that the risk of stroke was significantly increased in individuals with periodontitis in which the relative risk was 1.63 (95% confidence interval [CI] 1.25–2.00) [5]. Furthermore, a large cohort study has demonstrated that periodontal disease is associated with the incidence of cardioembolic and atherothrombotic stroke, and that regular dental care might decrease stroke risk [6]. Periodontitis is related to an increase in systemic inflammation markers through exposure to Gram-negative bacteria, which are implicated in the etiology of stroke [7]. While the treatment of stroke has improved remarkably, the management of periodontal disease may not only improve the clinical course in patients but could help prevent stroke altogether. However, in these studies periodontal diseases were diagnosed by only oral examination.

It was reported that serum IgG titers of a certain periodontal pathogen are considered to reflect its periodontal status [8]. Recently, by analyzing serum antibody titers, certain periodontal pathogens have been identified as risk factors for systemic diseases, such as ischemic stroke, coronary heart disease, non-alcoholic fatty liver disease, and Alzheimer's disease [9-12]. Especially regarding ischemic stroke, the serum antibody level of Prevotella (P.) intermedia was significantly higher in atherothrombotic stroke patients than in patients with no previous stroke [9]. However, there is no reported association between the serum IgG titers of periodontal pathogens and cerebral hemorrhage.

Comment 6: P. 6: Regarding the aim of the study at the end of the introduction – ‘Therefore, to understand how’ – could be rephrase for – ‘to determine if’ – since the outcome is about prognosis and not pathophysiology.

Response 6: We have revised the manuscript as per your suggestion.

Page 6, Lines 80-82

To determine if periodontal disease affects cerebral hemorrhage, we examined the relationship among serum IgG titers of periodontal pathogens, cerebral hemorrhage growth, and a 3-month outcome.

2. Materials and methods:

Comment 7: P. 6: It is not clear all the inclusion and exclusion criteria for participants (e.g. >18 years-old, etiologies stated as ‘others’ cerebral hemorrhage causes).

Response 7: We have revised the section on the enrollment criteria as follows:

Page 7, Lines 94-99

We included patients who were admitted within 7 days from onset, were aged ≥ 20 years, and for whom consent to participate in this study was obtained from the patient or their relatives. We excluded patients who could not undergo head computed tomography (CT) and magnetic resonance imaging (MRI). We also excluded the patients who were diagnosed with hemorrhagic infarction or trauma-induced hemorrhages.

Comment 8: P. 7: Please confirm if participants underwent head computed tomography and/or magnetic resonance at admission, and please clarify who performed the NIHSS assessment.

Response 8: Imaging analysis with head computed tomography (CT) and magnetic resonance imaging was performed in all patients for acute cerebral hemorrhage diagnosis. Accordingly, we added the following text:

Page 8, Lines 105-106

Two stroke specialists (SA and EI) evaluated the stroke severity and conscious level.

Comment 9: P. 7-8: It is worth to include the references for previously established definitions such as: ‘hypertension’ (e.g. American or European Hypertension Guidelines), ‘diabetes mellitus’, ‘dyslipidemia’, and ‘atrial fibrillation’; as there could be small differences on diagnostic criteria.

Response 9: We appreciate your wise advice. We have added citations and references to the appropriate reports and guidelines.

Page 8, Lines 107-110

Conscious level was evaluated with Glasgow coma scale. Comorbidities were defined according to a previous report [13] based on the Japanese hypertension, diabetes mellitus, dyslipidemia, atrial fibrillation, and chronic kidney disease guidelines.

Comment 10: P. 8: It is worth to explicitly state if the neuroimaging evaluation (i.e. hematoma volume measurements) remained blinded from the clinical assessment.

Response 10: We have added the following text to the revised manuscript:

Page 9, Line 128

The neuroimaging evaluation remained blinded from the clinical assessment.

Comment 11: P. 9: Regarding the determination of serum IgG titers, more detail about sampling and processing methods would be welcomed or they could be revisited as supplementary material.

Response 11: We have added a detailed method to the revised manuscript as follows:

Page 10, Lines 144-153

Bacterial antigen-coated wells were washed with phosphate-buffered saline with Tween (PBST); serum samples in PBST were then added to the wells. After incubation at 4°C overnight, the wells were washed with PBST and filled with alkaline phosphatase-conjugated goat anti-human IgG (gamma-chain specific, Abcam, Cambridge, MO) in PBST. After another incubation at 37°C for 2 hours, the wells were again washed with PBST, an aliquot of p-nitrophenylphosphate at 1 mg/mL (WAKO Pure Chemical Industries Ltd., Osaka, Japan) in 10% diethanolamine buffer was added to each well as a substrate, and incubation was performed at 37°C for 30 minutes. Optical density at 405 nm was measured using a microplate reader (iMark, Bio-Rad Laboratories Inc., Hercules, CA).

3. Discussion

Comment 12: P. 16, line 3: It is recommended to unified the format ‘(listed in Tables 1 & 2, and cerebral hematoma growth)’ instead it could be referred as ‘(listed in Tables. 1, 2 and 3)’.

Response 12: We have revised the format as you have suggested.

Comment 13: P. 16: Limitations should be updated, including comments regarding sample size and others addressed at the beginning of this review.

Response 13: We appreciate your suggestions. We had calculated the sample size and performed post hoc analysis. In addition, this study was performed by multicenter. However, there might exist some sampling bias as the reviewers mentioned. We described and added these sampling bias as limitations in the ‘Discussion’ section.

Page 26, Lines 279-291

There were some limitations to this study. First, there was the issue of sample size and sampling bias. In this study, the sample size was modest. Thus, despite the multicenter study design, sample size calculation, and post hoc analysis, some sampling bias might exist. The average age of patients and the numbers of patients with CAA, previous stroke, and taking antithrombotic drugs were also relatively high, whereas the total frequency of intracerebral hematoma growth was low. Regarding this low intracerebral hematoma growth frequency, the time from onset to admission might have affected the results. In this study, we included patients within 7 days from onset. However, all patients were considered during the acute phase, and the median time from onset to admission was 147 minutes. Time from onset to collection of blood sample also varied because of the study design. However, all patients were tested within a week from onset. Since periodontal disease is a chronic disease, mild variation might not affect the results.

4. Title

Comment 14: P. 1: Consider using a more accurate term such as ‘association’ or ‘relationship’ instead of ‘predict’. This would be more appropriate for the reader expectations.

Response 14: We have changed the title as per your suggestion.

Submitted filename: response_to_reviewer_submission.docx

Click here for additional data file.

12 Oct 2020

Serum IgG titers against periodontal pathogens are associated with cerebral hemorrhage growth and 3-month outcome

PONE-D-20-12711R1

Dear Dr. Hosomi,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Miguel A. Barboza, MD, MSc

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Authors have clarified the doubts in a timely manner. Comments have been answered as far as possible.

Reviewer #2: I was delighted to review the manuscript. It addresses an interesting major topic (i.e. intracerebral haemorrhage) from a non-conventional and deeply explored perspective (i.e. periodontal pathogens) and it gives place to potential future. The authors reasonably addressed prior comments to their paper. The manuscript is now easy to follow and the results and conclusions are coherent with the methodology. Finally, I would strongly suggest to the authors to continue exploring in further research this interesting association in a higher selective population with more rigorous inclusion and exclusion criteria.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

16 Oct 2020

PONE-D-20-12711R1

Serum IgG titers against periodontal pathogens are associated with cerebral hemorrhage growth and 3-month outcome

Dear Dr. Hosomi:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Miguel A. Barboza

Academic Editor

PLOS ONE