THE JOURNAL OF INFECTIOUS DISEASES. VOL. 138, NO.3. SEPTEMBER 1978 © 1978 by The University of Chicago. 0022-1899/78/3803-0018$00.75
Reactivation of Herpes Simplex Virus after Decompression of the Trigeminal Nerve Root George J. Pazin, Monto Ho, and Peter J. Jannetta
From the Departments of Medicine and Neurosurgery, School of Medicine, and the Department of Microbiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania
Development of cutaneous herpetic lesions following section of the trigeminal sensory root has been well documented [1-3]. Recently, a new operation, microneurosurgical decompression of the trigeminal nerve root, has been carried out to relieve intractable tic douloureux and atypical trigeminal neuralgia [4]. Compressing blood vessels are mierosurgically dissected from the trigeminal nerve root, and a tiny sponge is inserted between the nerve root and the vessel. No direct injury to the nerve occurs, but herpetic lesions have been noted postoperatively. A retrospective questionnaire revealed that -35% of patients Received for publication November 16, 1977, and in revised form March 8,1978. This study was presented in part at the Seventeenth Interscience Conference on Antimicrobial Agents and Chemotherapy, New York, New York, October 1977. This study was partially supported by grant no. 5ROI02953 from the National Institutes of Health. We acknowledge the cooperation of Drs. Robert H. Wilkins and David S. Zorub who permitted us to follow their patients, Ms. Leona Youngblood and Ms. Peggy Hosckswender for technical assistance, Drs. John A. Armstrong and George Tan for assistance and advice, and Ms. Betty Edwards for secretarialassistance. Please address requests for reprints to Dr. George J. Pazin, Department of Medicine, 968 Scaife Hall, University of Pittsburgh, Pittsburgh, Pennsylvania 15261.
with a history of herpes labialis developed herpetic lesions following this operation. The existence of patients in whom we might be able to predict reactivation of herpes simplex virus (HSV) following a specific, therapeutically useful, well-tolerated stimulus prompted this prospective investigation. Materials and Methods
Study population. Seventy patients were interviewed regarding a history of herpes labialis and were examined for herpetic lesions before surgery. A throat-wash (TW) culture and serum for a CF test for HSV were obtained. Fourteen patients were not followed postoperatively; the other 56 patients had adequate clinical and laboratory observations. TW cultures were obtained at various intervals during the postoperative period. All patients had at least one culture performed between days 2 and 10 after the operation. Patients were routinely examined for cutaneous herpetic lesions and oral lesions when cultures were obtained and sporadically by house staff between cultures. Throat culture method. Hanks' balanced salt solution (10 ml) with 0.5% gelatin was provided in sterile collection cups. Patients swished the so-
405
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
Reactivation of herpes simplex virus was prospectively studied in patients after microneurosurgical decompression of the trigeminal sensory root, a new operation for trigeminal neuralgia in which the nerve is not sectioned. Reactivation was detected in 28 (50%) of 56 patients. Virus was cultured from oropharyngeal secretions in 25 patients, and 21 patients developed cutaneous herpetic lesions. Seven patients had positive throat-wash (TW) cultures but did not develop lesions, and the converse occurred in three patients. Eighteen patients had both positive TW cultures and herpetic lesions. In eight of nine instances in which a sequence was determinable, TW cultures were positive before lesions developed. A history of recurrent herpes labialis was associated with a higher risk of developing reactivation postoperatively (59.4% vs. 31.6%, P < 0.05). These observations suggest that minimal stimulation or inapparent trauma to the trigeminal sensory root is sufficient to activate latent herpes simplex virus in humans. These patients provide unique opportunities to study immunologic responses and therapeutic measures.
406
Pazin, Ho, and [annetta
/ \
~ -.
pons
,
The operation is depicted in simplified fashion in figure 1. Vascular structures compressing the nerve varied, but an elongated loop of superior cerebellar artery was commonly adjacent to the nerve root entry zone. With the aid of a dissecting microscope at a magnification of X lOX 16, the vessel was dissected free and displaced from the nerve root. A tiny sponge was fashioned and inserted between the nerve and arterial loop. Microscopic dissection time was -15-30 min. Operations were not accompanied by diminution in facial sensation. A small area of lower lip trauma from the endotracheal tube was common on days 1 and 2 after the operation but bore no relation to subsequent labial herpetic lesions. Results
Evaluation before the operation. Of 70 patients interviewed before surgery, 42 (60%) had a history of herpes labialis. Sixty-six TW cultures were obtained before the operation, and three (4.5%) were positive in the absence of clinical herpes labialis. Two of the latter three patients had no history of herpes labialis. All three patients showed subsequent evidence of reactivation; two had negative TW cultures on day 1 after the operation, but all developed herpetic lesions and positive TW cultures on days 2-4. One patient had a resolving herpetic lesion before surgery. The TW culture of this patient was
superior cerebellar artery (looping course)
"I~
~~ .. ~~'~~' ,\...--
\
'. - '. ~
... -
.....;.,
~ ~ ~ .x .
pons
.~ --
superior cerebellar artery placed horizontally)
Figure 1. A simplified illustration of microneurosurgical decompression of the trigeminal sensory root in patients.
,( loop
~ ~V---
I
,~
" ',">...
--
pons
'Y--~ /. -. -~
,"'- iY
-- ,
"sponge"
interposed
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
lution in their mouths, gargled, and then expectorated it into the cup. Antibiotics (1 mg of gentamicin and 100 units of mycostatin) were added, and 0.5-0.6 ml was inoculated on duplicate monolayers of human fibroblasts (WI-3S, FS4, or locally derived), Vero cells, and primary monkey kidney cells in tube cultures, which were incubated at 37 C. Specimens not processed immediately were stored at -70 C. Cultures were observed for 14 days, and HSV was identified by typical CPE, which never developed later than the fourth day of observation. Isolates from eight patients were tested for HSV by fluorescent anti. body staining, and all specimens were strongly positive. Typical vesicular lesions on or about the lips were not cultured routinely, but vesicular lesions in atypical sites (chin, cheek, or away from the lips) were confirmed by culture. Serologic testing. A microtiter CF procedure, using HSV antigen, normal tissue antigen, and a positive control serum obtained from Microbiological Associates, Walkersville, Md., was performed. Serial twofold dilutions were tested, beginning at a dilution of l:S. A titer of ~1:S was considered positive. A negative control was included routinely. Operation. All patients received high dosages of corticosteroids (total equivalent of 90 mg of dexamethasone) before and after the operation over 72 hr. A retromastoid craniotomy was performed under general endotracheal anesthesia.
407
HSV after Trigeminal Decompression
Table 1. Results of oropharyngeal cultures for herpes simplex virus and appearance of herpetic lesions before and after trigeminal nerve root decompression in 28 patients with reactivated herpes labialis and 28 patients without reactivated infections. Oropharyngeal cultures* Patients with infections
Day Before operation After operation 1
2 3 4 5 6 7 8 9
10 Total after operation
Patients without infections
3/25 (12)
0/27
1/12 (8.3) 3/18 (17) 5/7 (71) 12/15 (80) 3{6 (50) 5/9 (56) 5{8 (63) 2/3 (67) 3/6 (50) 1/2 (50)
0/14 0/13 0/11 0/7 0/7 0/6 0/8 0/2 0/1 0/1
40/86 (47)
0/70
Le-
sionsf
o 1 6
7 1 4 1
o 1
o 21
*Data are given as number positive/number of cultures done that day (percentage). tData are number of patients according to first day of appearance of lesions. tOne lesion noted before surgery was in the healing phase. New vesicles appeared on day 3 after surgery.
and 4 after the operation was reflected in an abrupt increase in the percentage of positive TW cultures in the reactivation group to 70%-80%. Persistence of activation and onset in additional patients resulted in positivity of 50%-6070 of TW specimens from the reactivation group between days 5 and 10. Sequence of positive TW cultures and herpetic lesions in individual patients. Eighteen patients had both positive TW cultures and herpetic lesions, but a sequence of positivity was not determinable in nine of these patients. When one manifestation preceded the other, development of positive TW specimens occurred first in eight patients, whereas there was one instance of the converse. The site of shedding into the mouth was not identified despite careful examination. Relationship of side of lesion to side of operation. Lateralization of cutaneous herpetic lesions was evident in 16 patients and always occurred on the side of the operation. Three patients had central lesions, one had severe stoma-
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
negative before the operation but became positive on day 1 after operation, and new vesicles developed on day 3. From these observations it would appear that virus in oropharyngeal secretions or herpetic lesions before the operation augur well for development of herpes labialis after the operation. Postoperative follow-up. Of 56 patients followed postoperatively, 28 did not develop herpetic lesions, nor was HSV isolated on their TW cultures. Seventy TW specimens (2.5 per patient) were obtained for culture from these patients between days 1 and 10 after the operation. Twenty-eight patients (50%) developed evidence of reactivation. Twenty-five patients had positive TW cultures, and 21 developed herpetic lesions. Eighteen had both positive TW cultures and herpetic lesions. Seven patients had positive TW cultures but did not develop lesions, whereas the converse occurred in three patients. Eightyfive TW cultures (3.1 per patient) were performed after the operation in this group. Effect of history of herpes labialis. Twentytwo (59.4%) of 37 patients with a history of herpes labialis developed reactivated infections, whereas six (31.6%) of 19 patients without a history did so (P < 0.05). Effect of CF antibodies. Thirty-seven samples of serum were obtained before the operation. The CF test as done by the clinical immunopathology laboratory was not sensitive. Only four (1870) of 22 patients with proven postoperative reactivation had a positive CF test. However, the CF test did identify two patients with subsequent reactivation who had no history of herpes labialis. Time of reactivation. The proportion of TW specimens positive on each postoperative day and the first day on which herpetic lesions were noted are indicated in table 1. One of the TW specimens obtained on day 1 was positive. It came from the patient who had a healing cutaneous herpetic lesion, but a negative TW specimen, before the operation. Only three TW specimens obtained on day 2 after the operation were positive, and a single lesion was found. Again, one of the patients with positive TW specimens obtained on day 2 had been asymptomatically excreting HSV before surgery but had negative TW cultures one day after surgery. The frequent onset of reactivation on days 3
408
titis with bilateral lesions, and in one patient the site was not recorded. Postoperative events as possible stimuli for reactivation. Analysis of charts of 22 patients with reactivation and 22 without reactivation revealed no appreciable differences between groups with regard to total duration of surgery or postoperative fever, leukocytosis, or complications. Discussion
quently than does lesion development suggests that mucosal epithelium may be more susceptible to lysis or release of virus than cutaneous epithelium. Alternatively, the virus may have a shorter distance to travel to mucosal surfaces as discussed below. The earlier occurrence of positive TW specimens suggests the possibility that in some instances herpes labialis may be secondary to oropharyngeal infection, particularly if the lesions are not lateralized. Documentation of HSV reactivation in response to microneurosurgical manipulation of the trigeminal sensory root provides further insight into the pathogenesis of reactivation. Earlier investigators suggested that sectioning of the nerve root results in stimulation or damage of the ganglion cells, which leads to reactivation of virus that was possibly latent in the skin [2]. More recently, cocultivation recovery of HSV from human trigeminal ganglia without peripheral disease has lent support to the trigeminal ganglia as the site of latency [6-9]. Now it appears that mechanical manipulation or decompression of the trigeminal sensory root is sufficient to cause reactivation, and sectioning of the nerve is not necessary. Similar results were recently reported in a rabbit model by Nesburn et al. [10]. In addition to the self-evident differences in species being studied, our situation differs inasmuch as the sensory root rather than the ganglion was manipulated and our patients' initial infections occurred naturally and much earlier. The usual 48- to 96-hr period between surgery and detection of viral reactivation was similar to delays in development of cutaneous lesions that have been reported following trigeminal sensory root section. Presumably, this period represents time needed for activation of the virus in situ, migration of the virus from the ganglion to the periphery, and development of a lesion. The rate of movement of HSV in nerves in experimental models varies from 1.7 to 10 mmjhr [11-13]. Estimating 9 em from the ganglion to perioral skin, one can calculate that viral movement might reo quire 9-53 hr. A period of 48-72 hr was reo quired for lesion development in studies of HSV autoinoculation [14]. Our experience agrees with prior observations of reactivation following trigeminal sensory root surgery [2] inasmuch as lateralization of lesions, when present, always correlated with the side of surgery. This finding provides evidence that
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
In this prospective study, HSV infection was detected in 28 of 56 patients who underwent microneurosurgical decompression of the posterior root of the trigeminal nerve. Twenty-one patients (38%) developed cutaneous herpetic lesions within 10 days following surgery, a reactivation rate which is less than that (68%-94%) reported by other investigators after section of the same nerve [1-3], but substantial considering the minimal trauma to the nerve during the decompression operation. The more frequent manifestation of postoperative reactivation was the appearance of HSV in TW specimens. Previous reports of herpes labialis after sectioning of the trigeminal sensory root had not included TW cultures. Since the oral mucosa presumably was infected during the primary HSV infection, oropharyngeal excretion with reactivation was not surprising. We believe that the frequent isolations of HSV in TW specimens (40 [25.6%] of 156 postoperative specimens) represented true reactivation secondary to surgery because contamination could be disregarded when TW positivity preceded lesions or occurred in the absence of lesions (together, 54% of reactivation situations) and because spontaneous oropharyngeal activation is uncommon. Only 3.6% of 607 throat specimens cultured for HSV were positive in a prospective study of unprovoked reactivation of herpes labialis [5]. TW specimens may provide an opportunity to quantitate reactivation, and cultures before the operation are important for identification of patients who are asymptomatically reactivated before surgery. Our observation that three (4.5%) of 66 TW cultures obtained before the operation were positive without overt disease is similar to the rate of 2.2% found by Douglas and Couch [5]. The observation that the appearance of HSV in TW specimens occurs earlier and more fre-
Pazin, Ho, and [annetta
409
HSV after Trigeminal Decompression
References I. Carton, C. A., Kilbourne, E. D. Activation of latent herpes simplex by trigeminal sensory-root section. N. Engl. J. Med. 246:172-176,1952. 2. Carton, C. A. Effect of previous sensory loss on the appearance of herpes simplex following trigeminal sensory root section. J. Neurosurg. 10:463-468, 1953. 3. Ellison, S. A., Carton, C. A., Rose, H. M. Studies of recurrent herpes simplex infections following section of the trigeminal nerve. J. Infect. Dis. 105: 161-167, 1959. 4. Jannctta, P. J. Microsurgical approach to the trigeminal nerve for tic douloureux. In H. Krayenbuhl, R. F. Maspes, and W. H. Sweet [ed.], Progress in neurological surgery. Vol. 7. S. Karger, Basel, 1976, p.180-200. 5. Douglas, R. G., Jr., Couch, R. B. A prospective study of chronic herpes simplex virus infection and recurrent herpes labialis in humans. J. Immunol. 104: 289-295,1970. 6. Bastian, F. 0., Rabson, A. S., Yee, C. L., Tralka, T. S. Herpesvirus hominis: isolation from human trigeminal ganglion. Science 178:306-307, 1972. 7. Baringer, J. R., Swoveland, P. Recovery of herpessimplex virus from human trigeminal ganglions. N. Engl. J. Med. 288:648-650,1973. 8. Plummer, G. Isolation of herpesviruses from trigeminal ganglia of man, monkeys and cats. J. Infect. Dis. 128:345-348, 1973. 9. Forghani, B., Klassen, T., Baringer, J. R. Radioimmunoassay of herpes simplex virus antibody: correlation with ganglionic infection. J. Gen. Virol. 36:371-375, 1977. 10. Nesburn, A. B., Green, M. T., Radnoti, M., Walker, B. Reliable in vivo model for latent herpes simplex virus reactivation with peripheral virus shedding. Infec.lmmun. 15:772-775, 1977. II. Field, H. J., Hill, T. J. The pathogenesis of pseudorabies in mice: virus replication at the inoculation site and axonal uptake, J. Gen. Virol. 26:145-148, 1975. 12. Kristcnsson, K., Lycke, E., Sjostrand, J. The spread of herpes simplex virus in peripheral nerves. Acta Ncuropathol. (Berl.) 17:44-53,1971. 13. McCracken, R. M., McFerran, J. B., Dow, C. The neural spread of pseudorabies virus in calves. J. Gcn. Virol. 20:17-28,1973. 14. Blank, H., Haines, H. G. Experimental human reinfection with herpes simplex virus. J. Invest. Dermatol. 61:223-225,1973. 15. Stevens, J. G., Cook, M. L., Jordan, M. C. Reactivation of latent herpes simplex virus after pneumococcal pucumouia in mice. Infec. Immun, 11:635-639, 1975. 16. Underwood, G. E., Wced, S. D. Recurrent cutaneous herpes simplex in hairless mice. Infec. Immun. 10: 471-474, 1974.
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
manipulation of the sensory root was the principal stimulus for reactivation. Although passage of virus from the ganglion cells via intact sensory fibers to the periphery may occur, this route still has not been conclusively demonstrated in experimental models [15]. Since all patients in this study received high dosages of corticosteroids in conjunction with their surgery, we cannot determine the effect of steroids on reactivation. Prednisone therapy has been implicated in cutaneous HSV reactivation in experimental animals, but the effect was infrequent and variable [16]. We doubt that steroid therapy contributed significantly to reactivation in our investigation, but this possibility requires further study. Two potential preoperative risk factors, presence of antibody in serum and a history of herpetic lesions, were studied. CF tests as performed in our clinical laboratory lacked sensitivity and were not useful. It should be noted that in the detailed study by Ellison et al. of CF and neutralizing antibodies to HSV in relation to sectioning of the trigeminal sensory root [3], all patients had preexisting antibodies but there were no significant differences in titers between patients who developed herpes labialis after the operation and those who did not. There was also no correlation between titers before the operation and severity of herpetic ;teacti vation. In contrast, in our study a history of herpes labialis was effective in identifying populations at high (59.4%) and low (31.6%) risk for reactivation. This finding is consistent with the suggestion that herpes labialis was a reactivated rather than a de novo infection. However, other unknown host or agent factors such as differences in cell-mediated immunity may playa role in the increased risk of reactivation after surgery in patients with a history of herpes labialis. Reactivation of HSV in this study was usually benign clinically, but one patient developed severe stomatitis, and another developed symptomatic herpetic esophagi tis. Nevertheless, these generally healthy patients, in whom we can expect reactivation of latent virus following a precisely timed, therapeutically desired, well-tolerated stimulus, provide unusual opportunities to study reactivation of latent virus in humans as well as to investigate measures for prevention of this reactivation.
Reactivation of Herpes Simplex Virus after Decompression of the Trigeminal Nerve Root George J. Pazin, Monto Ho, and Peter J. Jannetta
From the Departments of Medicine and Neurosurgery, School of Medicine, and the Department of Microbiology, Graduate School of Public Health, University of Pittsburgh, Pittsburgh, Pennsylvania
Development of cutaneous herpetic lesions following section of the trigeminal sensory root has been well documented [1-3]. Recently, a new operation, microneurosurgical decompression of the trigeminal nerve root, has been carried out to relieve intractable tic douloureux and atypical trigeminal neuralgia [4]. Compressing blood vessels are mierosurgically dissected from the trigeminal nerve root, and a tiny sponge is inserted between the nerve root and the vessel. No direct injury to the nerve occurs, but herpetic lesions have been noted postoperatively. A retrospective questionnaire revealed that -35% of patients Received for publication November 16, 1977, and in revised form March 8,1978. This study was presented in part at the Seventeenth Interscience Conference on Antimicrobial Agents and Chemotherapy, New York, New York, October 1977. This study was partially supported by grant no. 5ROI02953 from the National Institutes of Health. We acknowledge the cooperation of Drs. Robert H. Wilkins and David S. Zorub who permitted us to follow their patients, Ms. Leona Youngblood and Ms. Peggy Hosckswender for technical assistance, Drs. John A. Armstrong and George Tan for assistance and advice, and Ms. Betty Edwards for secretarialassistance. Please address requests for reprints to Dr. George J. Pazin, Department of Medicine, 968 Scaife Hall, University of Pittsburgh, Pittsburgh, Pennsylvania 15261.
with a history of herpes labialis developed herpetic lesions following this operation. The existence of patients in whom we might be able to predict reactivation of herpes simplex virus (HSV) following a specific, therapeutically useful, well-tolerated stimulus prompted this prospective investigation. Materials and Methods
Study population. Seventy patients were interviewed regarding a history of herpes labialis and were examined for herpetic lesions before surgery. A throat-wash (TW) culture and serum for a CF test for HSV were obtained. Fourteen patients were not followed postoperatively; the other 56 patients had adequate clinical and laboratory observations. TW cultures were obtained at various intervals during the postoperative period. All patients had at least one culture performed between days 2 and 10 after the operation. Patients were routinely examined for cutaneous herpetic lesions and oral lesions when cultures were obtained and sporadically by house staff between cultures. Throat culture method. Hanks' balanced salt solution (10 ml) with 0.5% gelatin was provided in sterile collection cups. Patients swished the so-
405
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
Reactivation of herpes simplex virus was prospectively studied in patients after microneurosurgical decompression of the trigeminal sensory root, a new operation for trigeminal neuralgia in which the nerve is not sectioned. Reactivation was detected in 28 (50%) of 56 patients. Virus was cultured from oropharyngeal secretions in 25 patients, and 21 patients developed cutaneous herpetic lesions. Seven patients had positive throat-wash (TW) cultures but did not develop lesions, and the converse occurred in three patients. Eighteen patients had both positive TW cultures and herpetic lesions. In eight of nine instances in which a sequence was determinable, TW cultures were positive before lesions developed. A history of recurrent herpes labialis was associated with a higher risk of developing reactivation postoperatively (59.4% vs. 31.6%, P < 0.05). These observations suggest that minimal stimulation or inapparent trauma to the trigeminal sensory root is sufficient to activate latent herpes simplex virus in humans. These patients provide unique opportunities to study immunologic responses and therapeutic measures.
406
Pazin, Ho, and [annetta
/ \
~ -.
pons
,
The operation is depicted in simplified fashion in figure 1. Vascular structures compressing the nerve varied, but an elongated loop of superior cerebellar artery was commonly adjacent to the nerve root entry zone. With the aid of a dissecting microscope at a magnification of X lOX 16, the vessel was dissected free and displaced from the nerve root. A tiny sponge was fashioned and inserted between the nerve and arterial loop. Microscopic dissection time was -15-30 min. Operations were not accompanied by diminution in facial sensation. A small area of lower lip trauma from the endotracheal tube was common on days 1 and 2 after the operation but bore no relation to subsequent labial herpetic lesions. Results
Evaluation before the operation. Of 70 patients interviewed before surgery, 42 (60%) had a history of herpes labialis. Sixty-six TW cultures were obtained before the operation, and three (4.5%) were positive in the absence of clinical herpes labialis. Two of the latter three patients had no history of herpes labialis. All three patients showed subsequent evidence of reactivation; two had negative TW cultures on day 1 after the operation, but all developed herpetic lesions and positive TW cultures on days 2-4. One patient had a resolving herpetic lesion before surgery. The TW culture of this patient was
superior cerebellar artery (looping course)
"I~
~~ .. ~~'~~' ,\...--
\
'. - '. ~
... -
.....;.,
~ ~ ~ .x .
pons
.~ --
superior cerebellar artery placed horizontally)
Figure 1. A simplified illustration of microneurosurgical decompression of the trigeminal sensory root in patients.
,( loop
~ ~V---
I
,~
" ',">...
--
pons
'Y--~ /. -. -~
,"'- iY
-- ,
"sponge"
interposed
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
lution in their mouths, gargled, and then expectorated it into the cup. Antibiotics (1 mg of gentamicin and 100 units of mycostatin) were added, and 0.5-0.6 ml was inoculated on duplicate monolayers of human fibroblasts (WI-3S, FS4, or locally derived), Vero cells, and primary monkey kidney cells in tube cultures, which were incubated at 37 C. Specimens not processed immediately were stored at -70 C. Cultures were observed for 14 days, and HSV was identified by typical CPE, which never developed later than the fourth day of observation. Isolates from eight patients were tested for HSV by fluorescent anti. body staining, and all specimens were strongly positive. Typical vesicular lesions on or about the lips were not cultured routinely, but vesicular lesions in atypical sites (chin, cheek, or away from the lips) were confirmed by culture. Serologic testing. A microtiter CF procedure, using HSV antigen, normal tissue antigen, and a positive control serum obtained from Microbiological Associates, Walkersville, Md., was performed. Serial twofold dilutions were tested, beginning at a dilution of l:S. A titer of ~1:S was considered positive. A negative control was included routinely. Operation. All patients received high dosages of corticosteroids (total equivalent of 90 mg of dexamethasone) before and after the operation over 72 hr. A retromastoid craniotomy was performed under general endotracheal anesthesia.
407
HSV after Trigeminal Decompression
Table 1. Results of oropharyngeal cultures for herpes simplex virus and appearance of herpetic lesions before and after trigeminal nerve root decompression in 28 patients with reactivated herpes labialis and 28 patients without reactivated infections. Oropharyngeal cultures* Patients with infections
Day Before operation After operation 1
2 3 4 5 6 7 8 9
10 Total after operation
Patients without infections
3/25 (12)
0/27
1/12 (8.3) 3/18 (17) 5/7 (71) 12/15 (80) 3{6 (50) 5/9 (56) 5{8 (63) 2/3 (67) 3/6 (50) 1/2 (50)
0/14 0/13 0/11 0/7 0/7 0/6 0/8 0/2 0/1 0/1
40/86 (47)
0/70
Le-
sionsf
o 1 6
7 1 4 1
o 1
o 21
*Data are given as number positive/number of cultures done that day (percentage). tData are number of patients according to first day of appearance of lesions. tOne lesion noted before surgery was in the healing phase. New vesicles appeared on day 3 after surgery.
and 4 after the operation was reflected in an abrupt increase in the percentage of positive TW cultures in the reactivation group to 70%-80%. Persistence of activation and onset in additional patients resulted in positivity of 50%-6070 of TW specimens from the reactivation group between days 5 and 10. Sequence of positive TW cultures and herpetic lesions in individual patients. Eighteen patients had both positive TW cultures and herpetic lesions, but a sequence of positivity was not determinable in nine of these patients. When one manifestation preceded the other, development of positive TW specimens occurred first in eight patients, whereas there was one instance of the converse. The site of shedding into the mouth was not identified despite careful examination. Relationship of side of lesion to side of operation. Lateralization of cutaneous herpetic lesions was evident in 16 patients and always occurred on the side of the operation. Three patients had central lesions, one had severe stoma-
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
negative before the operation but became positive on day 1 after operation, and new vesicles developed on day 3. From these observations it would appear that virus in oropharyngeal secretions or herpetic lesions before the operation augur well for development of herpes labialis after the operation. Postoperative follow-up. Of 56 patients followed postoperatively, 28 did not develop herpetic lesions, nor was HSV isolated on their TW cultures. Seventy TW specimens (2.5 per patient) were obtained for culture from these patients between days 1 and 10 after the operation. Twenty-eight patients (50%) developed evidence of reactivation. Twenty-five patients had positive TW cultures, and 21 developed herpetic lesions. Eighteen had both positive TW cultures and herpetic lesions. Seven patients had positive TW cultures but did not develop lesions, whereas the converse occurred in three patients. Eightyfive TW cultures (3.1 per patient) were performed after the operation in this group. Effect of history of herpes labialis. Twentytwo (59.4%) of 37 patients with a history of herpes labialis developed reactivated infections, whereas six (31.6%) of 19 patients without a history did so (P < 0.05). Effect of CF antibodies. Thirty-seven samples of serum were obtained before the operation. The CF test as done by the clinical immunopathology laboratory was not sensitive. Only four (1870) of 22 patients with proven postoperative reactivation had a positive CF test. However, the CF test did identify two patients with subsequent reactivation who had no history of herpes labialis. Time of reactivation. The proportion of TW specimens positive on each postoperative day and the first day on which herpetic lesions were noted are indicated in table 1. One of the TW specimens obtained on day 1 was positive. It came from the patient who had a healing cutaneous herpetic lesion, but a negative TW specimen, before the operation. Only three TW specimens obtained on day 2 after the operation were positive, and a single lesion was found. Again, one of the patients with positive TW specimens obtained on day 2 had been asymptomatically excreting HSV before surgery but had negative TW cultures one day after surgery. The frequent onset of reactivation on days 3
408
titis with bilateral lesions, and in one patient the site was not recorded. Postoperative events as possible stimuli for reactivation. Analysis of charts of 22 patients with reactivation and 22 without reactivation revealed no appreciable differences between groups with regard to total duration of surgery or postoperative fever, leukocytosis, or complications. Discussion
quently than does lesion development suggests that mucosal epithelium may be more susceptible to lysis or release of virus than cutaneous epithelium. Alternatively, the virus may have a shorter distance to travel to mucosal surfaces as discussed below. The earlier occurrence of positive TW specimens suggests the possibility that in some instances herpes labialis may be secondary to oropharyngeal infection, particularly if the lesions are not lateralized. Documentation of HSV reactivation in response to microneurosurgical manipulation of the trigeminal sensory root provides further insight into the pathogenesis of reactivation. Earlier investigators suggested that sectioning of the nerve root results in stimulation or damage of the ganglion cells, which leads to reactivation of virus that was possibly latent in the skin [2]. More recently, cocultivation recovery of HSV from human trigeminal ganglia without peripheral disease has lent support to the trigeminal ganglia as the site of latency [6-9]. Now it appears that mechanical manipulation or decompression of the trigeminal sensory root is sufficient to cause reactivation, and sectioning of the nerve is not necessary. Similar results were recently reported in a rabbit model by Nesburn et al. [10]. In addition to the self-evident differences in species being studied, our situation differs inasmuch as the sensory root rather than the ganglion was manipulated and our patients' initial infections occurred naturally and much earlier. The usual 48- to 96-hr period between surgery and detection of viral reactivation was similar to delays in development of cutaneous lesions that have been reported following trigeminal sensory root section. Presumably, this period represents time needed for activation of the virus in situ, migration of the virus from the ganglion to the periphery, and development of a lesion. The rate of movement of HSV in nerves in experimental models varies from 1.7 to 10 mmjhr [11-13]. Estimating 9 em from the ganglion to perioral skin, one can calculate that viral movement might reo quire 9-53 hr. A period of 48-72 hr was reo quired for lesion development in studies of HSV autoinoculation [14]. Our experience agrees with prior observations of reactivation following trigeminal sensory root surgery [2] inasmuch as lateralization of lesions, when present, always correlated with the side of surgery. This finding provides evidence that
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
In this prospective study, HSV infection was detected in 28 of 56 patients who underwent microneurosurgical decompression of the posterior root of the trigeminal nerve. Twenty-one patients (38%) developed cutaneous herpetic lesions within 10 days following surgery, a reactivation rate which is less than that (68%-94%) reported by other investigators after section of the same nerve [1-3], but substantial considering the minimal trauma to the nerve during the decompression operation. The more frequent manifestation of postoperative reactivation was the appearance of HSV in TW specimens. Previous reports of herpes labialis after sectioning of the trigeminal sensory root had not included TW cultures. Since the oral mucosa presumably was infected during the primary HSV infection, oropharyngeal excretion with reactivation was not surprising. We believe that the frequent isolations of HSV in TW specimens (40 [25.6%] of 156 postoperative specimens) represented true reactivation secondary to surgery because contamination could be disregarded when TW positivity preceded lesions or occurred in the absence of lesions (together, 54% of reactivation situations) and because spontaneous oropharyngeal activation is uncommon. Only 3.6% of 607 throat specimens cultured for HSV were positive in a prospective study of unprovoked reactivation of herpes labialis [5]. TW specimens may provide an opportunity to quantitate reactivation, and cultures before the operation are important for identification of patients who are asymptomatically reactivated before surgery. Our observation that three (4.5%) of 66 TW cultures obtained before the operation were positive without overt disease is similar to the rate of 2.2% found by Douglas and Couch [5]. The observation that the appearance of HSV in TW specimens occurs earlier and more fre-
Pazin, Ho, and [annetta
409
HSV after Trigeminal Decompression
References I. Carton, C. A., Kilbourne, E. D. Activation of latent herpes simplex by trigeminal sensory-root section. N. Engl. J. Med. 246:172-176,1952. 2. Carton, C. A. Effect of previous sensory loss on the appearance of herpes simplex following trigeminal sensory root section. J. Neurosurg. 10:463-468, 1953. 3. Ellison, S. A., Carton, C. A., Rose, H. M. Studies of recurrent herpes simplex infections following section of the trigeminal nerve. J. Infect. Dis. 105: 161-167, 1959. 4. Jannctta, P. J. Microsurgical approach to the trigeminal nerve for tic douloureux. In H. Krayenbuhl, R. F. Maspes, and W. H. Sweet [ed.], Progress in neurological surgery. Vol. 7. S. Karger, Basel, 1976, p.180-200. 5. Douglas, R. G., Jr., Couch, R. B. A prospective study of chronic herpes simplex virus infection and recurrent herpes labialis in humans. J. Immunol. 104: 289-295,1970. 6. Bastian, F. 0., Rabson, A. S., Yee, C. L., Tralka, T. S. Herpesvirus hominis: isolation from human trigeminal ganglion. Science 178:306-307, 1972. 7. Baringer, J. R., Swoveland, P. Recovery of herpessimplex virus from human trigeminal ganglions. N. Engl. J. Med. 288:648-650,1973. 8. Plummer, G. Isolation of herpesviruses from trigeminal ganglia of man, monkeys and cats. J. Infect. Dis. 128:345-348, 1973. 9. Forghani, B., Klassen, T., Baringer, J. R. Radioimmunoassay of herpes simplex virus antibody: correlation with ganglionic infection. J. Gen. Virol. 36:371-375, 1977. 10. Nesburn, A. B., Green, M. T., Radnoti, M., Walker, B. Reliable in vivo model for latent herpes simplex virus reactivation with peripheral virus shedding. Infec.lmmun. 15:772-775, 1977. II. Field, H. J., Hill, T. J. The pathogenesis of pseudorabies in mice: virus replication at the inoculation site and axonal uptake, J. Gen. Virol. 26:145-148, 1975. 12. Kristcnsson, K., Lycke, E., Sjostrand, J. The spread of herpes simplex virus in peripheral nerves. Acta Ncuropathol. (Berl.) 17:44-53,1971. 13. McCracken, R. M., McFerran, J. B., Dow, C. The neural spread of pseudorabies virus in calves. J. Gcn. Virol. 20:17-28,1973. 14. Blank, H., Haines, H. G. Experimental human reinfection with herpes simplex virus. J. Invest. Dermatol. 61:223-225,1973. 15. Stevens, J. G., Cook, M. L., Jordan, M. C. Reactivation of latent herpes simplex virus after pneumococcal pucumouia in mice. Infec. Immun, 11:635-639, 1975. 16. Underwood, G. E., Wced, S. D. Recurrent cutaneous herpes simplex in hairless mice. Infec. Immun. 10: 471-474, 1974.
Downloaded from http://jid.oxfordjournals.org/ at University of Regina on September 7, 2015
manipulation of the sensory root was the principal stimulus for reactivation. Although passage of virus from the ganglion cells via intact sensory fibers to the periphery may occur, this route still has not been conclusively demonstrated in experimental models [15]. Since all patients in this study received high dosages of corticosteroids in conjunction with their surgery, we cannot determine the effect of steroids on reactivation. Prednisone therapy has been implicated in cutaneous HSV reactivation in experimental animals, but the effect was infrequent and variable [16]. We doubt that steroid therapy contributed significantly to reactivation in our investigation, but this possibility requires further study. Two potential preoperative risk factors, presence of antibody in serum and a history of herpetic lesions, were studied. CF tests as performed in our clinical laboratory lacked sensitivity and were not useful. It should be noted that in the detailed study by Ellison et al. of CF and neutralizing antibodies to HSV in relation to sectioning of the trigeminal sensory root [3], all patients had preexisting antibodies but there were no significant differences in titers between patients who developed herpes labialis after the operation and those who did not. There was also no correlation between titers before the operation and severity of herpetic ;teacti vation. In contrast, in our study a history of herpes labialis was effective in identifying populations at high (59.4%) and low (31.6%) risk for reactivation. This finding is consistent with the suggestion that herpes labialis was a reactivated rather than a de novo infection. However, other unknown host or agent factors such as differences in cell-mediated immunity may playa role in the increased risk of reactivation after surgery in patients with a history of herpes labialis. Reactivation of HSV in this study was usually benign clinically, but one patient developed severe stomatitis, and another developed symptomatic herpetic esophagi tis. Nevertheless, these generally healthy patients, in whom we can expect reactivation of latent virus following a precisely timed, therapeutically desired, well-tolerated stimulus, provide unusual opportunities to study reactivation of latent virus in humans as well as to investigate measures for prevention of this reactivation.
