G Model
ARTICLE IN PRESS
JCV 2984 1–7
Journal of Clinical Virology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Clinical Virology journal homepage: www.elsevier.com/locate/jcv
Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients
1
2
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Q1
Uluhan Sili a,∗ , Abdurrahman Kaya a,1 , Ali Mert b , HSV Encephalitis Study Group, Resat Ozaras c , Kenan Midilli d , Sait Albayram e , Gulay Kenangil f , Onat Demirci g , Mahir Kapmaz h , Kadriye Kart Yasar i , Bircan Unal Kayaaslan j , Sehnaz Ozyavuz Alp k , Mustafa Ozgun Yılmaz l a
Department of Infectious Diseases and Clinical Microbiology, Istanbul University Cerrahpasa School of Medicine, Istanbul, Turkey Department of Internal Medicine, Istanbul Medipol University School of Medicine, Istanbul, Turkey c Cerrahpasa Medical School, Department of Infectious Diseases and Clinical Microbiology, Turkey d Cerrahpasa Medical School, Department of Microbiology and Clinical Microbiology, Section of Molecular Microbiology, Turkey e Cerrahpasa Medical School, Department of Radiology, Section of Neuroradiology, Turkey f Sisli Etfal Education and Research Hospital, Department of Neurology, Turkey g Cerrahpasa Medical School, Department of Neurology, Turkey h Istanbul University School of Medicine (Capa), Department of Infectious Diseases and Clinical Microbiology, Turkey i Haseki Education and Research Hospital, Department of Infectious Diseases and Clinical Microbiology, Turkey j Ankara Numune Education and Research Hospital, Department of Infectious Diseases and Clinical Microbiology, Turkey k Hacettepe University School of Medicine, Department of Medicine, Section of Infectious Diseases, Turkey l Gazi University School of Medicine, Department of Infectious Diseases and Clinical Microbiology, Turkey b
19
20 31
a r t i c l e
i n f o
a b s t r a c t
21 22 23 24 25
Article history: Received 4 January 2014 Received in revised form 24 February 2014 Accepted 15 March 2014
26 27 28 29 30
Keywords: Herpes simplex virus viral encephalitis prognostic factors
Background: Herpes simplex virus (HSV) is one the most common causes of sporadic encephalitis worldwide. Objective: We aimed to determine clinical characteristics and prognosis of HSV encephalitis (HSVE) cases reviewed retrospectively from several collaborating centers. Study design: We searched hospital archives of the last 10 years for patients with HSVE diagnosis, i.e. clinical presentation compatible with encephalitis and brain involvement on magnetic resonance imaging (MRI) or detection of HSV DNA in the cerebrospinal fluid by polymerase chain reaction (PCR). Clinical characteristics were noted and patients were phone-interviewed. HSVE cases were grouped and analyzed as proven and probable, based on virological confirmation by PCR. Univariate and multivariate analyses were used to determine factors associated with prognosis. Results: A total of 106 patients (63 males; mean age, 44 years; range, 18–83 years) were included. Most common symptoms were changes in mental status, fever, headache, and seizure. HSV PCR was positive in 69% of patients tested, while brain involvement was detected on MRI in 95%. Acyclovir was started mostly within five days of main symptom and continued for ≥14 days. Case fatality rate was 8%, while 69% of patients recovered with sequelae. Favorable prognosis was observed in 73% of patients. Multivariate analysis identified the duration of disease before hospital admission (odds ratio (OR) = 1.24) and the extent of brain involvement on MRI at the time of admission (OR = 37.22) as two independent risk factors associated with poor prognosis. Conclusions: Although HSVE fatality regressed considerably with acyclovir treatment, many patients survive with sequelae. Our results emphasize the importance of early diagnosis and prompt treatment of HSVE. © 2014 Published by Elsevier B.V. 32
Abbreviations: CI, confidence interval; CSF, cerebrospinal fluid; HSV, herpes simplex virus; HSVE, herpes simplex virus encephalitis; MRI, magnetic resonance imaging; OR, odds ratio; PCR, polymerase chain reaction. ∗ Corresponding author. Current address: Marmara Universitesi, Pendik Egitim ve Arastirma Hastanesi, Enfeksiyon Hastaliklari Anabilim Dali, Fevzi Cakmak Mah. Mimar Sinan Cad. No: 41, Ust Kaynarca, Pendik, 34896 Istanbul, Turkey. Tel.: +90 505 746 5265; fax: +90 216 625 4790; mobile: +90 216 625 4693. E-mail addresses: [email protected], [email protected] (U. Sili). 1 Address: Suleymaniye Kadin ve Cocuk Hastaliklari Egitim ve Arastırma Hastanesi, Enfeksiyon Hastaliklari, Telsiz Mah. Balikli Kazli Cesme Yolu No: 3, Zeytinburnu, 34020 Istanbul, Turkey. Tel.: +90 506 611 3328; fax: +90 212 416 9814; mobile: +90 0212 664 5355. http://dx.doi.org/10.1016/j.jcv.2014.03.010 1386-6532/© 2014 Published by Elsevier B.V.
Please cite this article in press as: Sili U, et al. Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients. J Clin Virol (2014), http://dx.doi.org/10.1016/j.jcv.2014.03.010
G Model JCV 2984 1–7
U. Sili et al. / Journal of Clinical Virology xxx (2014) xxx–xxx
2 33
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
52
53 54 55
56
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
ARTICLE IN PRESS
1. Background Herpes simplex virus encephalitis (HSVE) is one of the most common causes of sporadic necrotizing encephalitis globally [1,2]. HSVE is characterized by fever, headache, mental status changes, seizures, and focal neurological deficits that develop acutely [3]. HSVE is caused by invasion of the brain parenchyma by HSV through primary infection or reactivation of the latent virus. Encephalitis typically involves the temporal lobe and is best visualized by magnetic resonance imaging (MRI) of the brain [4]. The detection of HSV DNA in cerebrospinal fluid (CSF) using polymerase chain reaction (PCR) confirms the diagnosis of HSVE [5]. HSVE-related fatality rate and morbidity were around 70% and 50%, respectively, before the development of effective antiviral therapy [3,6,7]. With the introduction of acyclovir in mid-1980s [8,9], HSVE became a much more treatable disease, and consequently HSVE-related one-year fatality rate declined to 5–15% [2,10]. Despite this, neuropsychiatric sequelae are still common [10]. Poor prognosis was shown to be associated with delays in the diagnosis and treatment of HSVE [11–14].
calculated the performance of this method. The sensitivity and specificity of predicting poor outcome using the last observation method was 86% and 94%, respectively. Thus, we included these 29 patients in our analyses. We scored the level of morbidity according to the publication by Whitley et al. with modifications [9]: • No sequelae: patient states that his/her health is exactly like the pre-encephalitic period. • Mild sequelae: patient has subjective minor disability compared to pre-encephalitic stage such as memory impairment or decrease in attention span; not on anti-epileptic treatment; able to work and function autonomously. • Moderate sequelae: patient has subjective major disability compared to pre-encephalitic stage or is on anti-epileptic treatment; mostly self-sufficient in daily routine; states that his/her life has significant differences compared to the pre-encephalitic stage; cannot sustain a regular job. • Severe sequelae: major neuropsychiatric disability or chronic care patient; Kluver-Bucy syndrome; need constant help for daily routine.
93 94 95 96 97 98
99 100 101 102 103 104 105 106 107 108 109 110 111 112
2. Objective We aimed to determine the diagnosis, treatment, and prognosisrelated clinical features of HSVE cases reviewed retrospectively from several collaborating centers in Turkey. 3. Study design Archives of infectious disease and neurology departments of 17 hospitals were searched retrospectively for adult HSVE cases. To be included in the study, a patient should have presented with the clinical characteristics of encephalitis (changes in mental status, abnormal behavior, speech disturbances, epilepsy) alongside typical involvement of brain on MRI and/or PCR detection of HSV DNA in CSF. Due to the retrospective nature of the study spanning several years, PCR methodology used to detect HSV DNA varied between the centers. A total of 106 patients diagnosed in a time frame of 10 years, from 2001 to 2012, constituted our HSVE group. All cases were above 18 years of age. Demographic data, duration and characteristics of complaints related to encephalitis, physical examination findings, laboratory values at the time of admission, PCR detection of HSV DNA in the CSF, electroencephalography, brain computed tomography (CT), and MRI were retrieved. For each case, medical notes were extensively reviewed and symptoms deemed to be related to encephalitis were recorded. Symptoms were classified as “main symptom”, when it caused emergency department admission and “first symptom”, when it was present from the beginning of the illness. Treatment related characteristics such as acyclovir duration, start of treatment in relation to main and first symptoms, and development of nephrotoxicity (defined as absolute increase in serum creatinine of ≥0.3 mg/dL within 48 h of treatment) were noted. Patients with somnolence, stupor, or coma at admission were regarded as having severely depressed level of consciousness for the purpose of analysis. MRI brain involvement was categorized as unilateral, bilateral, or extensive. Outcome assessment was conducted on 97 (92%) patients, as there was no follow-up information on nine patients. To determine the level of morbidity, we conducted a standardized telephone interview with the patients and/or their relatives, at least six months after discharge. Twenty-nine cases (30%) could not be reached for a telephone interview. For these cases, the last observation on their medical files was extrapolated as the outcome. To verify this approach, we extrapolated the outcome of interviewed patients using the last observation in their medical files and
PCR positive and PCR negative/not done cases were grouped as “proven” and “probable”, respectively, and compared for any statistically significant differences. To identify the factors that determine the outcome, patients with fatality or severe sequelae were classified as “poor prognosis”, while patients with no or mildto-moderate sequelae were classified as “favorable prognosis.” Univariate analyses were carried out for comparing poor and favorable prognosis within proven, probable and total groups. Mann–Whitney U test was used for comparing continuous variables since the data did not follow normal distribution. Chi-square or Fisher’s exact tests were used for comparing categorical variables. Multivariate analysis was carried out with covariates that showed a p value of 11,000/mm3 . b Neutrophilia, >7500/mm3 . c Hyponatremia, Na 50% of total cell count. e Hypoglycorrhachia, CSF glucose 45 mg/dL.
3
Table 3 Diagnostic test results (n = 106). Test
Requested, n (%)
PCR 80 (75) 66 (62) EEG CT 58 (55) 106 (100) MRI Localization of lesion on MRI (n = 98)a Right temporal only Left temporal only Bilateral temporal Extensive
Positive result, n (%) 55 (69) 57 (86) 30 (52) 101 (95) 24 (24) 37 (38) 20 (20) 17 (17)
CT, computed tomography; EEG, electroencephalography; MRI, magnetic resonance imaging; PCR, polymerase chain reaction. a Localization of lesion was missing from the files of three patients with a positive MRI.
included lymphocytic pleocytosis in 86%, hypoglycorrhachia in 25%, and elevated protein levels in 70% of patients. CSF was acellular in 15% of patients. MRI was the most common diagnostic method. Brain involvement was detectable by MRI in 101 patients (95%, Table 3). All MRI positive cases showed temporal lobe involvement typical of HSV encephalitis. All MRI negative cases (n = 5) had clinical presentations compatible with encephalitis and positive HSV PCR. HSV PCR was found positive in 55 of 80 cases tested (69%, Table 3). Of the HSV PCR positive cases, 91% were HSV type 1. Mean duration between the first symptom and the start of acyclovir was 6.15 ± 5 days (range, 1–30 days), while mean duration between the main symptom and the start of acyclovir was 2.38 ± 2.6 days (range, 1–20 days). Appropriate treatment was started within five days of main symptom in 91% of cases and within eight days of first symptom in 83% of cases. There was no discernible delay between emergency admission and start of treatment suggesting that in tertiary centers doctors were able to recognize encephalitis in time and start appropriate treatment within 24–48 h. Acyclovir treatment duration was 18.7 ± 5.4 days (range, 10–42 days). Except for eight patients (8%), all patients received ≥14 days of treatment. In 32% of patients, nephrotoxicity was noted. Patients developing nephrotoxicity were significantly older (50.6 ± 17.3 years vs. 38.8 ± 13.1 years, p = 0.005). Nephrotoxicity was reversible in all cases with increased hydration. As there was no follow-up information in nine patients, analysis of prognostic factors was carried out with 97 patients (92%, Table 4). Six-month fatality rate was 8% (n = 8). While 23% of patients had complete recovery, 69% had sequelae. Even though the level of sequelae in most of the patients was mild-to-moderate, 19% had severe sequelae. In general, majority of the cases (73%) resulted in favorable prognosis.
Table 4 Sequelae and prognosis of HSVE cases (n = 97).a n (%) No sequelae Sequelae Mild Moderate Severe Fatality rate Prognosis Favorableb Poorc
22 (23) 67 (69) 31 (32) 18 (19) 18 (19) 8 (8) 71 (73) 26 (27)
a As there was no follow-up information for nine patients, outcome assessment was performed on 97 patients. b Favorable prognosis included no sequelae or mild-moderate sequelae. c Poor prognosis included severe sequelae or fatality.
Please cite this article in press as: Sili U, et al. Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients. J Clin Virol (2014), http://dx.doi.org/10.1016/j.jcv.2014.03.010
157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
G Model JCV 2984 1–7
ARTICLE IN PRESS U. Sili et al. / Journal of Clinical Virology xxx (2014) xxx–xxx
4
Table 5 Comparison of proven (PCR positive) and probable (PCR negative or not done) HSVE cases. Proven cases (n = 53)
Probable cases (n = 44)
p
Gender (male/female) Age (years), mean ± SD (range)
34 (64)/19 (36) 44 ± 16
24 (55)/20 (46) 45 ± 16
0.337 0.928
Abnormal behavior Speech disturbance Loss of consciousness Confusion/disorientation Seizure Headache Nausea/vomiting
38 (73) 31 (62) 25 (49) 47 (92) 29 (56) 39 (83) 24 (56)
23 (61) 22 (58) 23 (55) 31 (80) 24 (60) 27 (73) 21 (60)
0.208 0.697 0.581 0.080 0.684 0.267 0.710
Fever Fever duration (days) Main symptom duration (days) First symptom duration (days)
43 (81) 4.7 ± 3.1 2.2 ± 2.0 6.5 ± 5.6
32 (76) 3.6 ± 3.3 2.5 ± 3.3 4.7 ± 3.3
0.557 0.053 0.824 0.044
Severely depressed level of consciousnessa Meningeal irritation signs Focal neurological signs
17 (32) 15 (29) 14 (28)
17 (39) 16 (39) 13 (31)
0.500 0.332 0.700
Leukocyte (/mm3 ) CSF leukocyte (/mm3 ) EEG positivity CT positivity
10,743 ± 3041 110 ± 133 29 (88) 16 (50)
10,317 ± 3151 180 ± 227 24 (89) 13 (52)
0.411 0.608 1.000 0.881
Extent of brain involvement in MRI and prognosis Unilateral Bilateral Extensive Duration of hospitalization (days) Treatment duration Nephrotoxicity
30 (64) 8 (17) 9 (19) 23 ± 12 19 ± 5 12 (31)
24 (57) 10 (24) 8 (19) 23 ± 10 17 ± 4 8 (33)
14 (29) 34 (71) 18 (53) 9 (27) 7 (21) 5 (9)
8 (20) 33 (81) 13 (39) 9 (27) 11 (33) 3 (7)
0.293 0.293
41 (77) 12 (22)
30 (68) 14 (32)
0.310
No sequelae Sequelae Mild Moderate Severe Fatality rate Prognosis Favorable Poor
0.716 0.855 0.189 0.832
0.432 0.725
Data are given as mean ± standard deviation or n (%). a Severely depressed level of consciousness included somnolence, stupor and coma. 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
We grouped PCR negative and not done cases as probable HSVE and compared to the virologically proven PCR positive cases (Table 5). This analysis revealed that there were no statistically significant differences between probable and proven HSVE cases in terms of demographic data, clinical presentation, laboratory/imaging findings, and prognosis, except for the first symptom duration. This result suggested that PCR negative cases within the probable group could be false negative. As there were no repeat PCR results for these cases, we cannot verify this assumption. However, we compared the duration of disease when CSF sampling was performed between PCR negative and positive cases to determine whether timing of CSF sampling could account for PCR negativity. In PCR negative cases lumbar puncture was performed about 2.5 days earlier than PCR positive cases, when disease duration was measured from the first symptom (4.2 ± 2.7 days vs. 6.7 ± 5.6 days, p = 0.019). To determine the factors affecting prognosis, we grouped patients as having favorable and poor prognosis within total, proven, and probable HSVE groups, and performed univariate analysis (Table 6). Longer duration of main and first symptoms before emergency admission, longer hospitalization, presence of abnormal behavior, meningeal irritation signs, severely depressed level of consciousness, and extensive brain involvement determined by MRI at the time of admission were factors significantly associated with poor prognosis for the total group. Other than first and main symptom durations and duration of hospitalization, there were no variables that became significant with the addition of probable
cases to the proven ones. Moreover, for the statistically significant variables in proven cases, addition of probable cases yielded higher significance. Thus, we combined patients with proven and probable HSVE in order to run multivariate analysis. Logistic regression analysis revealed that longer duration of main symptom before hospital admission and extensive brain involvement on MRI at admission were two independent risk factors associated with poor prognosis (Table 7). 5. Discussion The most significant factor affecting the HSVE outcome is starting effective antiviral treatment as early as possible [7]. Empirical acyclovir treatment is recommended when encephalitis is suspected [15,16]. Despite improvements in diagnosis and treatment, about 30% of patients with HSVE result in poor prognosis [12,17]. In the latest studies, one-year fatality rate is reported as 5–14%, while a staggering 89% are discharged with persistent neurological symptoms, suggesting room for improvement [2,10]. Ours is the first clinical series investigating the management of HSVE cases in Turkey. Among presenting symptoms were mental status changes, fever, headache, and seizure. CSF analyses revealed that there were mononuclear pleocytosis, mild increase in protein levels, and normal glucose levels in many patients. Cerebral involvement was detected by MRI in almost all cases. Ninety percent of patients received effective treatment within five days of the start of main symptom. Atypical findings included no fever (24%),
Please cite this article in press as: Sili U, et al. Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients. J Clin Virol (2014), http://dx.doi.org/10.1016/j.jcv.2014.03.010
216 217 218 219 220 221 222 223
224
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240
G Model
ARTICLE IN PRESS
JCV 2984 1–7
U. Sili et al. / Journal of Clinical Virology xxx (2014) xxx–xxx
5
Table 6 Univariate analysis with total (n = 97), virologically proven (n = 53) and probable (n = 44) HSVE cases.
Total
Proven
Probable
Favorable prognosis
Poor prognosis
p
Favorable prognosis
Poor prognosis
p
Favorable prognosis
Poor prognosis
p
44 (62) / 27 (38)
14 (54)/12(46)
0.470
27 (66) /14 (34)
7 (58) / 5 (42)
0.736
17 (57)/13 (43)
7(50) /7 (50)
0.679
Age (years)
43.9±15
45.6±17.4
0.687
45.7±16.1
48.8±19.9
0.726
43.5±13.5
57.0±19.5
0.319
Abnormal behavior
40 (60)
21 (91)
0.005
26 (65)
12 (100)
0.023
14 (52)
9 (82)
0.145
Speech disturbance
38 (56)
15 (71)
0.229
23 (58)
8 (80)
0.282
15 (56)
7 (64)
0.729
Loss of consciousness
33 (48)
15 (62)
0.215
20 (50)
5 (46)
0.789
13 (49)
10 (77)
0.053
Confusion/ disorientation
56 (84)
22 (96)
0.284
35 (90)
12 (100)
0.561
21 (75)
10 (91)
0.400
Seizure
39 (56)
14 (64)
0.512
24 (59)
5 (46)
0.507
15 (52)
9 (82)
0.148
Headache
50 (78)
16 (80)
1
32 (84)
7 (78)
0.639
18 (70)
9 (82)
0.688
Nausea/ vomiting
33 (54)
12 (71)
0.224
17 (49)
7 (88)
0.059
16 (62)
5 (56)
1.000
Fever
53 (76)
22 (88)
0.196
32 (78)
11 (92)
0.423
21 (72)
11 (85)
0.466
Fever duration (days)
3.9±2.9
5.1±3.8
0.333
4.8±3.5
4.2±3.0
0.601
2.6±1.6
4.0±2.4
0.054
1.94±1.63
3.4±4.3
0.042
1.73±1.54
2.00±1.41
0.085
1.94±1.65
1.75±0.50
0.261
Parameter
Gender (male/female)
Main symptom duration (days)
Data are given as mean ± standard deviation or n (%). Severely depressed level of consciousness included somnolence, stupor and coma.
a
241 242 243 244 245 246 247 248 249
normal consciousness (10%), meningeal irritation signs (32%), and hypoglycorrhachia (25%). Six-month fatality rate was 8% and morbidity of varying degrees was observed in 69% of cases. Full recovery was seen in only 23% of cases. Favorable prognosis was observed in 73% of our patients, similar to published reports [11,12,17]. PCR detection of HSV DNA in CSF is the method of choice in the diagnosis of HSVE [5,18]. In our series, PCR was negative in 31% of the tested cases. However, PCR can be false negative at early stages
of the disease and has to be repeated 3–7 days later in patients with suspected encephalitis [19–22]. None of our PCR negative cases were retested possibly because result of initial PCR was usually received within 3–4 weeks (slow turnaround time), by which time most patients had already improved clinically or been discharged so that a repeat sampling could not be done. When we compared the time of CSF sampling between PCR positive and negative cases, we found that patients with negative HSV PCR in CSF had significantly shorter duration of disease suggesting PCR negativity could be due
Please cite this article in press as: Sili U, et al. Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients. J Clin Virol (2014), http://dx.doi.org/10.1016/j.jcv.2014.03.010
250 251 252 253 254 255 256 257 258
G Model
ARTICLE IN PRESS
JCV 2984 1–7
U. Sili et al. / Journal of Clinical Virology xxx (2014) xxx–xxx
6
Table 7 Logistic regression to determine factors associated with prognosis (n = 97). OR Duration of main symptom, days Brain involvement in MRI Right or left temporal only Bilateral temporal Extensive
259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313
1.24 1 2.05 37.22
95% CI 1.06–1.53
0.41–10.34 7.09–195.47
p 0.045
0.39
ARTICLE IN PRESS
JCV 2984 1–7
Journal of Clinical Virology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of Clinical Virology journal homepage: www.elsevier.com/locate/jcv
Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients
1
2
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Q1
Uluhan Sili a,∗ , Abdurrahman Kaya a,1 , Ali Mert b , HSV Encephalitis Study Group, Resat Ozaras c , Kenan Midilli d , Sait Albayram e , Gulay Kenangil f , Onat Demirci g , Mahir Kapmaz h , Kadriye Kart Yasar i , Bircan Unal Kayaaslan j , Sehnaz Ozyavuz Alp k , Mustafa Ozgun Yılmaz l a
Department of Infectious Diseases and Clinical Microbiology, Istanbul University Cerrahpasa School of Medicine, Istanbul, Turkey Department of Internal Medicine, Istanbul Medipol University School of Medicine, Istanbul, Turkey c Cerrahpasa Medical School, Department of Infectious Diseases and Clinical Microbiology, Turkey d Cerrahpasa Medical School, Department of Microbiology and Clinical Microbiology, Section of Molecular Microbiology, Turkey e Cerrahpasa Medical School, Department of Radiology, Section of Neuroradiology, Turkey f Sisli Etfal Education and Research Hospital, Department of Neurology, Turkey g Cerrahpasa Medical School, Department of Neurology, Turkey h Istanbul University School of Medicine (Capa), Department of Infectious Diseases and Clinical Microbiology, Turkey i Haseki Education and Research Hospital, Department of Infectious Diseases and Clinical Microbiology, Turkey j Ankara Numune Education and Research Hospital, Department of Infectious Diseases and Clinical Microbiology, Turkey k Hacettepe University School of Medicine, Department of Medicine, Section of Infectious Diseases, Turkey l Gazi University School of Medicine, Department of Infectious Diseases and Clinical Microbiology, Turkey b
19
20 31
a r t i c l e
i n f o
a b s t r a c t
21 22 23 24 25
Article history: Received 4 January 2014 Received in revised form 24 February 2014 Accepted 15 March 2014
26 27 28 29 30
Keywords: Herpes simplex virus viral encephalitis prognostic factors
Background: Herpes simplex virus (HSV) is one the most common causes of sporadic encephalitis worldwide. Objective: We aimed to determine clinical characteristics and prognosis of HSV encephalitis (HSVE) cases reviewed retrospectively from several collaborating centers. Study design: We searched hospital archives of the last 10 years for patients with HSVE diagnosis, i.e. clinical presentation compatible with encephalitis and brain involvement on magnetic resonance imaging (MRI) or detection of HSV DNA in the cerebrospinal fluid by polymerase chain reaction (PCR). Clinical characteristics were noted and patients were phone-interviewed. HSVE cases were grouped and analyzed as proven and probable, based on virological confirmation by PCR. Univariate and multivariate analyses were used to determine factors associated with prognosis. Results: A total of 106 patients (63 males; mean age, 44 years; range, 18–83 years) were included. Most common symptoms were changes in mental status, fever, headache, and seizure. HSV PCR was positive in 69% of patients tested, while brain involvement was detected on MRI in 95%. Acyclovir was started mostly within five days of main symptom and continued for ≥14 days. Case fatality rate was 8%, while 69% of patients recovered with sequelae. Favorable prognosis was observed in 73% of patients. Multivariate analysis identified the duration of disease before hospital admission (odds ratio (OR) = 1.24) and the extent of brain involvement on MRI at the time of admission (OR = 37.22) as two independent risk factors associated with poor prognosis. Conclusions: Although HSVE fatality regressed considerably with acyclovir treatment, many patients survive with sequelae. Our results emphasize the importance of early diagnosis and prompt treatment of HSVE. © 2014 Published by Elsevier B.V. 32
Abbreviations: CI, confidence interval; CSF, cerebrospinal fluid; HSV, herpes simplex virus; HSVE, herpes simplex virus encephalitis; MRI, magnetic resonance imaging; OR, odds ratio; PCR, polymerase chain reaction. ∗ Corresponding author. Current address: Marmara Universitesi, Pendik Egitim ve Arastirma Hastanesi, Enfeksiyon Hastaliklari Anabilim Dali, Fevzi Cakmak Mah. Mimar Sinan Cad. No: 41, Ust Kaynarca, Pendik, 34896 Istanbul, Turkey. Tel.: +90 505 746 5265; fax: +90 216 625 4790; mobile: +90 216 625 4693. E-mail addresses: [email protected], [email protected] (U. Sili). 1 Address: Suleymaniye Kadin ve Cocuk Hastaliklari Egitim ve Arastırma Hastanesi, Enfeksiyon Hastaliklari, Telsiz Mah. Balikli Kazli Cesme Yolu No: 3, Zeytinburnu, 34020 Istanbul, Turkey. Tel.: +90 506 611 3328; fax: +90 212 416 9814; mobile: +90 0212 664 5355. http://dx.doi.org/10.1016/j.jcv.2014.03.010 1386-6532/© 2014 Published by Elsevier B.V.
Please cite this article in press as: Sili U, et al. Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients. J Clin Virol (2014), http://dx.doi.org/10.1016/j.jcv.2014.03.010
G Model JCV 2984 1–7
U. Sili et al. / Journal of Clinical Virology xxx (2014) xxx–xxx
2 33
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
52
53 54 55
56
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
ARTICLE IN PRESS
1. Background Herpes simplex virus encephalitis (HSVE) is one of the most common causes of sporadic necrotizing encephalitis globally [1,2]. HSVE is characterized by fever, headache, mental status changes, seizures, and focal neurological deficits that develop acutely [3]. HSVE is caused by invasion of the brain parenchyma by HSV through primary infection or reactivation of the latent virus. Encephalitis typically involves the temporal lobe and is best visualized by magnetic resonance imaging (MRI) of the brain [4]. The detection of HSV DNA in cerebrospinal fluid (CSF) using polymerase chain reaction (PCR) confirms the diagnosis of HSVE [5]. HSVE-related fatality rate and morbidity were around 70% and 50%, respectively, before the development of effective antiviral therapy [3,6,7]. With the introduction of acyclovir in mid-1980s [8,9], HSVE became a much more treatable disease, and consequently HSVE-related one-year fatality rate declined to 5–15% [2,10]. Despite this, neuropsychiatric sequelae are still common [10]. Poor prognosis was shown to be associated with delays in the diagnosis and treatment of HSVE [11–14].
calculated the performance of this method. The sensitivity and specificity of predicting poor outcome using the last observation method was 86% and 94%, respectively. Thus, we included these 29 patients in our analyses. We scored the level of morbidity according to the publication by Whitley et al. with modifications [9]: • No sequelae: patient states that his/her health is exactly like the pre-encephalitic period. • Mild sequelae: patient has subjective minor disability compared to pre-encephalitic stage such as memory impairment or decrease in attention span; not on anti-epileptic treatment; able to work and function autonomously. • Moderate sequelae: patient has subjective major disability compared to pre-encephalitic stage or is on anti-epileptic treatment; mostly self-sufficient in daily routine; states that his/her life has significant differences compared to the pre-encephalitic stage; cannot sustain a regular job. • Severe sequelae: major neuropsychiatric disability or chronic care patient; Kluver-Bucy syndrome; need constant help for daily routine.
93 94 95 96 97 98
99 100 101 102 103 104 105 106 107 108 109 110 111 112
2. Objective We aimed to determine the diagnosis, treatment, and prognosisrelated clinical features of HSVE cases reviewed retrospectively from several collaborating centers in Turkey. 3. Study design Archives of infectious disease and neurology departments of 17 hospitals were searched retrospectively for adult HSVE cases. To be included in the study, a patient should have presented with the clinical characteristics of encephalitis (changes in mental status, abnormal behavior, speech disturbances, epilepsy) alongside typical involvement of brain on MRI and/or PCR detection of HSV DNA in CSF. Due to the retrospective nature of the study spanning several years, PCR methodology used to detect HSV DNA varied between the centers. A total of 106 patients diagnosed in a time frame of 10 years, from 2001 to 2012, constituted our HSVE group. All cases were above 18 years of age. Demographic data, duration and characteristics of complaints related to encephalitis, physical examination findings, laboratory values at the time of admission, PCR detection of HSV DNA in the CSF, electroencephalography, brain computed tomography (CT), and MRI were retrieved. For each case, medical notes were extensively reviewed and symptoms deemed to be related to encephalitis were recorded. Symptoms were classified as “main symptom”, when it caused emergency department admission and “first symptom”, when it was present from the beginning of the illness. Treatment related characteristics such as acyclovir duration, start of treatment in relation to main and first symptoms, and development of nephrotoxicity (defined as absolute increase in serum creatinine of ≥0.3 mg/dL within 48 h of treatment) were noted. Patients with somnolence, stupor, or coma at admission were regarded as having severely depressed level of consciousness for the purpose of analysis. MRI brain involvement was categorized as unilateral, bilateral, or extensive. Outcome assessment was conducted on 97 (92%) patients, as there was no follow-up information on nine patients. To determine the level of morbidity, we conducted a standardized telephone interview with the patients and/or their relatives, at least six months after discharge. Twenty-nine cases (30%) could not be reached for a telephone interview. For these cases, the last observation on their medical files was extrapolated as the outcome. To verify this approach, we extrapolated the outcome of interviewed patients using the last observation in their medical files and
PCR positive and PCR negative/not done cases were grouped as “proven” and “probable”, respectively, and compared for any statistically significant differences. To identify the factors that determine the outcome, patients with fatality or severe sequelae were classified as “poor prognosis”, while patients with no or mildto-moderate sequelae were classified as “favorable prognosis.” Univariate analyses were carried out for comparing poor and favorable prognosis within proven, probable and total groups. Mann–Whitney U test was used for comparing continuous variables since the data did not follow normal distribution. Chi-square or Fisher’s exact tests were used for comparing categorical variables. Multivariate analysis was carried out with covariates that showed a p value of 11,000/mm3 . b Neutrophilia, >7500/mm3 . c Hyponatremia, Na 50% of total cell count. e Hypoglycorrhachia, CSF glucose 45 mg/dL.
3
Table 3 Diagnostic test results (n = 106). Test
Requested, n (%)
PCR 80 (75) 66 (62) EEG CT 58 (55) 106 (100) MRI Localization of lesion on MRI (n = 98)a Right temporal only Left temporal only Bilateral temporal Extensive
Positive result, n (%) 55 (69) 57 (86) 30 (52) 101 (95) 24 (24) 37 (38) 20 (20) 17 (17)
CT, computed tomography; EEG, electroencephalography; MRI, magnetic resonance imaging; PCR, polymerase chain reaction. a Localization of lesion was missing from the files of three patients with a positive MRI.
included lymphocytic pleocytosis in 86%, hypoglycorrhachia in 25%, and elevated protein levels in 70% of patients. CSF was acellular in 15% of patients. MRI was the most common diagnostic method. Brain involvement was detectable by MRI in 101 patients (95%, Table 3). All MRI positive cases showed temporal lobe involvement typical of HSV encephalitis. All MRI negative cases (n = 5) had clinical presentations compatible with encephalitis and positive HSV PCR. HSV PCR was found positive in 55 of 80 cases tested (69%, Table 3). Of the HSV PCR positive cases, 91% were HSV type 1. Mean duration between the first symptom and the start of acyclovir was 6.15 ± 5 days (range, 1–30 days), while mean duration between the main symptom and the start of acyclovir was 2.38 ± 2.6 days (range, 1–20 days). Appropriate treatment was started within five days of main symptom in 91% of cases and within eight days of first symptom in 83% of cases. There was no discernible delay between emergency admission and start of treatment suggesting that in tertiary centers doctors were able to recognize encephalitis in time and start appropriate treatment within 24–48 h. Acyclovir treatment duration was 18.7 ± 5.4 days (range, 10–42 days). Except for eight patients (8%), all patients received ≥14 days of treatment. In 32% of patients, nephrotoxicity was noted. Patients developing nephrotoxicity were significantly older (50.6 ± 17.3 years vs. 38.8 ± 13.1 years, p = 0.005). Nephrotoxicity was reversible in all cases with increased hydration. As there was no follow-up information in nine patients, analysis of prognostic factors was carried out with 97 patients (92%, Table 4). Six-month fatality rate was 8% (n = 8). While 23% of patients had complete recovery, 69% had sequelae. Even though the level of sequelae in most of the patients was mild-to-moderate, 19% had severe sequelae. In general, majority of the cases (73%) resulted in favorable prognosis.
Table 4 Sequelae and prognosis of HSVE cases (n = 97).a n (%) No sequelae Sequelae Mild Moderate Severe Fatality rate Prognosis Favorableb Poorc
22 (23) 67 (69) 31 (32) 18 (19) 18 (19) 8 (8) 71 (73) 26 (27)
a As there was no follow-up information for nine patients, outcome assessment was performed on 97 patients. b Favorable prognosis included no sequelae or mild-moderate sequelae. c Poor prognosis included severe sequelae or fatality.
Please cite this article in press as: Sili U, et al. Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients. J Clin Virol (2014), http://dx.doi.org/10.1016/j.jcv.2014.03.010
157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188
G Model JCV 2984 1–7
ARTICLE IN PRESS U. Sili et al. / Journal of Clinical Virology xxx (2014) xxx–xxx
4
Table 5 Comparison of proven (PCR positive) and probable (PCR negative or not done) HSVE cases. Proven cases (n = 53)
Probable cases (n = 44)
p
Gender (male/female) Age (years), mean ± SD (range)
34 (64)/19 (36) 44 ± 16
24 (55)/20 (46) 45 ± 16
0.337 0.928
Abnormal behavior Speech disturbance Loss of consciousness Confusion/disorientation Seizure Headache Nausea/vomiting
38 (73) 31 (62) 25 (49) 47 (92) 29 (56) 39 (83) 24 (56)
23 (61) 22 (58) 23 (55) 31 (80) 24 (60) 27 (73) 21 (60)
0.208 0.697 0.581 0.080 0.684 0.267 0.710
Fever Fever duration (days) Main symptom duration (days) First symptom duration (days)
43 (81) 4.7 ± 3.1 2.2 ± 2.0 6.5 ± 5.6
32 (76) 3.6 ± 3.3 2.5 ± 3.3 4.7 ± 3.3
0.557 0.053 0.824 0.044
Severely depressed level of consciousnessa Meningeal irritation signs Focal neurological signs
17 (32) 15 (29) 14 (28)
17 (39) 16 (39) 13 (31)
0.500 0.332 0.700
Leukocyte (/mm3 ) CSF leukocyte (/mm3 ) EEG positivity CT positivity
10,743 ± 3041 110 ± 133 29 (88) 16 (50)
10,317 ± 3151 180 ± 227 24 (89) 13 (52)
0.411 0.608 1.000 0.881
Extent of brain involvement in MRI and prognosis Unilateral Bilateral Extensive Duration of hospitalization (days) Treatment duration Nephrotoxicity
30 (64) 8 (17) 9 (19) 23 ± 12 19 ± 5 12 (31)
24 (57) 10 (24) 8 (19) 23 ± 10 17 ± 4 8 (33)
14 (29) 34 (71) 18 (53) 9 (27) 7 (21) 5 (9)
8 (20) 33 (81) 13 (39) 9 (27) 11 (33) 3 (7)
0.293 0.293
41 (77) 12 (22)
30 (68) 14 (32)
0.310
No sequelae Sequelae Mild Moderate Severe Fatality rate Prognosis Favorable Poor
0.716 0.855 0.189 0.832
0.432 0.725
Data are given as mean ± standard deviation or n (%). a Severely depressed level of consciousness included somnolence, stupor and coma. 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215
We grouped PCR negative and not done cases as probable HSVE and compared to the virologically proven PCR positive cases (Table 5). This analysis revealed that there were no statistically significant differences between probable and proven HSVE cases in terms of demographic data, clinical presentation, laboratory/imaging findings, and prognosis, except for the first symptom duration. This result suggested that PCR negative cases within the probable group could be false negative. As there were no repeat PCR results for these cases, we cannot verify this assumption. However, we compared the duration of disease when CSF sampling was performed between PCR negative and positive cases to determine whether timing of CSF sampling could account for PCR negativity. In PCR negative cases lumbar puncture was performed about 2.5 days earlier than PCR positive cases, when disease duration was measured from the first symptom (4.2 ± 2.7 days vs. 6.7 ± 5.6 days, p = 0.019). To determine the factors affecting prognosis, we grouped patients as having favorable and poor prognosis within total, proven, and probable HSVE groups, and performed univariate analysis (Table 6). Longer duration of main and first symptoms before emergency admission, longer hospitalization, presence of abnormal behavior, meningeal irritation signs, severely depressed level of consciousness, and extensive brain involvement determined by MRI at the time of admission were factors significantly associated with poor prognosis for the total group. Other than first and main symptom durations and duration of hospitalization, there were no variables that became significant with the addition of probable
cases to the proven ones. Moreover, for the statistically significant variables in proven cases, addition of probable cases yielded higher significance. Thus, we combined patients with proven and probable HSVE in order to run multivariate analysis. Logistic regression analysis revealed that longer duration of main symptom before hospital admission and extensive brain involvement on MRI at admission were two independent risk factors associated with poor prognosis (Table 7). 5. Discussion The most significant factor affecting the HSVE outcome is starting effective antiviral treatment as early as possible [7]. Empirical acyclovir treatment is recommended when encephalitis is suspected [15,16]. Despite improvements in diagnosis and treatment, about 30% of patients with HSVE result in poor prognosis [12,17]. In the latest studies, one-year fatality rate is reported as 5–14%, while a staggering 89% are discharged with persistent neurological symptoms, suggesting room for improvement [2,10]. Ours is the first clinical series investigating the management of HSVE cases in Turkey. Among presenting symptoms were mental status changes, fever, headache, and seizure. CSF analyses revealed that there were mononuclear pleocytosis, mild increase in protein levels, and normal glucose levels in many patients. Cerebral involvement was detected by MRI in almost all cases. Ninety percent of patients received effective treatment within five days of the start of main symptom. Atypical findings included no fever (24%),
Please cite this article in press as: Sili U, et al. Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients. J Clin Virol (2014), http://dx.doi.org/10.1016/j.jcv.2014.03.010
216 217 218 219 220 221 222 223
224
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240
G Model
ARTICLE IN PRESS
JCV 2984 1–7
U. Sili et al. / Journal of Clinical Virology xxx (2014) xxx–xxx
5
Table 6 Univariate analysis with total (n = 97), virologically proven (n = 53) and probable (n = 44) HSVE cases.
Total
Proven
Probable
Favorable prognosis
Poor prognosis
p
Favorable prognosis
Poor prognosis
p
Favorable prognosis
Poor prognosis
p
44 (62) / 27 (38)
14 (54)/12(46)
0.470
27 (66) /14 (34)
7 (58) / 5 (42)
0.736
17 (57)/13 (43)
7(50) /7 (50)
0.679
Age (years)
43.9±15
45.6±17.4
0.687
45.7±16.1
48.8±19.9
0.726
43.5±13.5
57.0±19.5
0.319
Abnormal behavior
40 (60)
21 (91)
0.005
26 (65)
12 (100)
0.023
14 (52)
9 (82)
0.145
Speech disturbance
38 (56)
15 (71)
0.229
23 (58)
8 (80)
0.282
15 (56)
7 (64)
0.729
Loss of consciousness
33 (48)
15 (62)
0.215
20 (50)
5 (46)
0.789
13 (49)
10 (77)
0.053
Confusion/ disorientation
56 (84)
22 (96)
0.284
35 (90)
12 (100)
0.561
21 (75)
10 (91)
0.400
Seizure
39 (56)
14 (64)
0.512
24 (59)
5 (46)
0.507
15 (52)
9 (82)
0.148
Headache
50 (78)
16 (80)
1
32 (84)
7 (78)
0.639
18 (70)
9 (82)
0.688
Nausea/ vomiting
33 (54)
12 (71)
0.224
17 (49)
7 (88)
0.059
16 (62)
5 (56)
1.000
Fever
53 (76)
22 (88)
0.196
32 (78)
11 (92)
0.423
21 (72)
11 (85)
0.466
Fever duration (days)
3.9±2.9
5.1±3.8
0.333
4.8±3.5
4.2±3.0
0.601
2.6±1.6
4.0±2.4
0.054
1.94±1.63
3.4±4.3
0.042
1.73±1.54
2.00±1.41
0.085
1.94±1.65
1.75±0.50
0.261
Parameter
Gender (male/female)
Main symptom duration (days)
Data are given as mean ± standard deviation or n (%). Severely depressed level of consciousness included somnolence, stupor and coma.
a
241 242 243 244 245 246 247 248 249
normal consciousness (10%), meningeal irritation signs (32%), and hypoglycorrhachia (25%). Six-month fatality rate was 8% and morbidity of varying degrees was observed in 69% of cases. Full recovery was seen in only 23% of cases. Favorable prognosis was observed in 73% of our patients, similar to published reports [11,12,17]. PCR detection of HSV DNA in CSF is the method of choice in the diagnosis of HSVE [5,18]. In our series, PCR was negative in 31% of the tested cases. However, PCR can be false negative at early stages
of the disease and has to be repeated 3–7 days later in patients with suspected encephalitis [19–22]. None of our PCR negative cases were retested possibly because result of initial PCR was usually received within 3–4 weeks (slow turnaround time), by which time most patients had already improved clinically or been discharged so that a repeat sampling could not be done. When we compared the time of CSF sampling between PCR positive and negative cases, we found that patients with negative HSV PCR in CSF had significantly shorter duration of disease suggesting PCR negativity could be due
Please cite this article in press as: Sili U, et al. Herpes simplex virus encephalitis: Clinical manifestations, diagnosis and outcome in 106 adult patients. J Clin Virol (2014), http://dx.doi.org/10.1016/j.jcv.2014.03.010
250 251 252 253 254 255 256 257 258
G Model
ARTICLE IN PRESS
JCV 2984 1–7
U. Sili et al. / Journal of Clinical Virology xxx (2014) xxx–xxx
6
Table 7 Logistic regression to determine factors associated with prognosis (n = 97). OR Duration of main symptom, days Brain involvement in MRI Right or left temporal only Bilateral temporal Extensive
259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313
1.24 1 2.05 37.22
95% CI 1.06–1.53
0.41–10.34 7.09–195.47
p 0.045
0.39
