Uhthoff’s Symptom in Optic Neuritis: Relationship to Magnetic Resonance Imaging and Development of Multiple Sclerosis Georgina Bates Scholl, MB, BChir, Hu-Shien Song, MD, and Shirley H. Wray, MD, PhD, FRCP
Eighty-one patients with a first attack of isolated optic neuritis, 40 with Uhthoff’s symptom (Group 1)and 41 without (Group 21, were studied. All had a neurovisual examination, 74 of 81 patients had the pattern visual evoked potential recorded at rest, and 43 had magnetic resonance imaging brain scans. The pattern visual evoked potential PlOO latency was prolonged, Group 1 with a mean of 136 f 19 msec, Group 2 with a mean of 131 +- 19 msec (control subjects, 102 5 msec; n = 84), and the PlOO amplitude was reduced, without significant difference between the groups. Abnormal magnetic resonance imaging scans were present in significantly more patients in Group 1 ( p < 0.025). Treatment of optic neuritis with corticosteroids had no effect on the evolution or duration of Uhthoff’s symptom. Overall, 35 of 81 (43%) patients, followed for a mean of 3.5 years, developed multiple sclerosis. The incidence was significantly greater in Group 1 (p < 0.01). Uhthoff‘s symptom also correlated with a higher incidence of recurrent optic neuritis. We conclude that Uhthoff‘s symptom is a prognostic indicator for the early development of multiple sclerosis.
*
Scholl GB, Song H-S, Wray SH. Uhthoff’s symptom in optic neuritis: relationship to magnetic resonance imaging and development of multiple sclerosis. Ann Neurol 1991;30:180- 184
Uhthoff s symptom of intermittent transient blurring of vision on exertion has been described in multiple sclerosis (MS), isolated idiopathic optic neuritis, and in other optic neuropathies El-51. The symptom may also be provoked by emotional stress, temperature change, menstruation, increased illumination, eating, drinkmg, and smoking [G, 71. An inverse Uhthoffs symptom of improved vision has also been reported 18-10]. The pathophysiology of Uhthoff s symptom is unknown, although a reversible conduction block in demyelinated nerve fibers secondary to an increase in body temperature {11-141, or to changes in blood electrolyte levels or p H [l5, 161, is believed to play a role. The present study was designed to determine the incidence of Uhthoff s symptom in patients with a first attack of isolated optic neuritis and to compare the visual signs, the results of magnetic resonance imaging (MRI) of the brain, and the pattern visual evoked potentials (PVEP) of patients with and without Uhthoff s symptom. We chose to look at the PVEP at rest to see whether the neuropathy was accompanied by a delay in the Pl00 wave of equal magnitude in patiems with and without Uhthoffs symptom and if the range of
prolonged latency fell into the range consistent with a demyelinating process. The frequency of an abnormal MRI scan characterized by multifocal white matter lesions, in patients with clinically isolated optic neuritis, has been reported between 62 and 66% 117-131, but to date no study has compared the incidence of abnormal MRI scans in patients with and without Uhthoffs symptom. By looking at the incidence of an abnormal MRI scan, in conjunction with long-term follow-up for the development of clinical MS, we aimed to determine if the presence of Uhthoffs symptom could be predictive for the conversion of isolated idiopathic optic neuritis to MS.
From the Unit for Neurovisual Disorders, Department of Neurology, Massachusetts General Hospital, Boston, MA.
Address correspondence to Dr Scholl, Unit for Neurovisual Disorders, ACC 83?, Massachusetts General Hospital, 15 Parkman Street, MA 02’14’
Received Dec 12, 1990, and in revised form Feb 11, 1991. Accepted
Patients and Methods
Patients Eighty-one unselected patients with a first attack of acute idiopathic optic neuritis between February 1980 and December 1988 were studied in the Unit for Neurovisual Disorders at Massachusetts General Hospital (Boston, MA). None had a prior diagnosis of MS nor symptoms suggestive of demyelination. All patients were asked if they experienced temporary changes in vision with recovery back to baseline and, if so, what factors provoked the episodes. Seventy-five patients
for publication Feb 19, 1991.
180 Copyright 0 1991 by the American Neurological Association
had unilateral optic neuritis and 6 bilateral. There were 50 females and 3 1 males, aged 9 to 5 5 years; the mean age of females was 30 years and the mean age of males was 31.5 years. The patients were divided into the following two groups: Group 1, patients with Uhthoffs symptom, and Group 2, patients without Uhthoffs symptom. In Group 1, there were 40 patients, that is, 17 males and 23 females, aged 15 to 50 years, with a mean age of 30.9 years; in Group 2, there were 41 patients, that is, 14 males and 27 females, aged 9 to 55 years, with a mean age of 30.2 years. The PVEP was recorded in 84 normal control subjects with no history of eye disease, Uhthoffs symptom, or significant systemic disease, and a normal neuro-ophthalmic examination. There were 35 males aged 10 to 58 years, with a mean age of 30.7 years, and 49 females aged 8 to 74 years, with a mean age of 36.9 years.
Methods All patients had a neuro-ophthalmic examination and a workup to establish the diagnosis of acute idiopathic optic neuritis. Snellen visual acuity (VA) was recorded and expressed as a decimal (20/20 equivalent to 1.0). Follow-up permitted documentation of the worst VA and the best final VA. Fifty-eight patients, 33 in Group 1 and 25 in Group 2, also had a neurological examination to detect signs of MS. Additional studies to help establish a diagnosis of MS included somatosensory evoked responses, brainstem auditory evoked responses, and an examination of the cerebrospinal fluid (CSF). The clinical criteria of Perkin and Rose [20] for the diagnosis of MS in patients who have had optic neuritis were used for the clinical classification of MS into definite, probable, or possible. The additional category of laboratorysupported definite MS (criteria of Poser and coworkers 121)) was included for those patients with isolated optic neuritis and an abnormal CSF consistent with demyelination. Patients with an abnormal MRI scan but no other evidence of MS apart from optic neuritis were assigned to a fifth category of MRI-supported MS (criteria of Paty and colleagues 1191). The incidence of MS was calculated separately using these three diagnostic systems. A second attack of optic neuritis was not used in this study as a criterion for the diagnosis of MS. Patiems were followed for a mean of 3.5 years (Group 1 mean, 38 months; Group 2 mean, 45 months). Noncontrast MRI scans of the brain were obtained in 43 patients, that is, 23 in Group 1 and 20 in Group 2. 1'1 and T2-weighted images in the axial, sagittal, and coronal plains were obtained. Scans were interpreted as strongly suggestive of MS if there were three or more white matter lesions of high intensity on T2-weighted images, each measuring greater than 3 mm in diameter 1191. The PVEP was recorded in 35 of 40 patients in Group 1 and in 39 of 41 in Group 2 using methods reported elsewhere 122). All recordings were made during or after recovery from the initial attack of optic neuritis and before any subsequent attack. The latency and amplitude of the first major negative (N70) and positive (P100) peaks were measured. Check size 50' was chosen as the most likely to detect a difference in latency or amplitude in the two groups of patients. Responses are more frequently absent to smaller check sizes in patients with optic neuritis and visual loss.
Statistical analysis of the data was performed to look for significant correlations between the presence of Uhthoff s symptom and the other variables studied. Probability values were calculated by Fisher's exact test, x2 test, or Spearman correlation as appropriate and as indicated in the results.
Results Uhthoffs symptom was experienced by 40 of 81 (49.5%) patients with idiopathic optic neuritis at some time in the course of the attack and by 13 of 8 1 (16%) patients within 2 weeks of the onset of optic neuritis. In 1 patient, it commenced 6 years after the onset of optic neuritis, but most commonly, the symptom began within 2 months of the onset (23 of 40 patients). In 3 patients with early onset Uhthoff s symptom, there was a long period free of the symptom ( 3 , 14, and 32 months) followed by return of the symptom without a recurrent attack of optic neuritis. Episodes often continued over several months o r years. The frequency of episodes varied from sporadic to several episodes per day. Some patients noted an increase in frequency in the summer months. The provoking factors of Uhthoff s symptom are listed in Table 1. Exertion was the most common (21 of 40 patients), then hot baths or showers (11 of 40 patients) and hot weather (11 of 40 patients). T h e typical attack described by most of our patients was blurring of vision in the affected eye, which was usually generalized but was sometimes more pronounced centrally, coming o n after approximately 5 to 20 minutes of exposure to the provoking factor. A few patients noted color desaturation. Typically, vision recovered to its previous level over a variable period after resting or moving away from heat, commonly within 5 to 60 minutes, but sometimes taking u p to 24 hours. An athlete reported bilateral blurring of vision after 15 minutes of running, which after resting would remain blurred for 2 0 minutes, then recover to normal over the next 10 minutes. Only 1 patient experienced an inverse Uhthoffs symptom with improvement in vision under conditions of extreme stress at administrative meetings. The sensitivity to provocation of
Table 1 Factor5 Provoking Uhthaffs Symptom I
Factor
Patients (n)
Percentage
Physical exertion Hot bath or shower Hot weather Stress, anxiety, anger Tired, end of day Hot food or drink Cooking Other specific activity"
21 11 11 5 4
52.5 27.5 27.5 12.5 10 7.5
3 2 3
5 7.5
"Working a cash register, playing the trumpet, and reading (1 patient each).
Scholl et al: Uhthoff's Symptom
181
Table 2. Sarnmty of Clinical Data
Patients (n) Worst acuity at onset (mean) Worst acuity 50.3 (%) Best recovered visual acuity (mean) Recovered to 1.0 Treated with steroids (%) Having single attack (%) Having multiple attacks (%) Incidence of MS (%) Clinical criteria" Clinical and CSFb Clinical, CSF, and MRI' Incidence of MS' by sex (%) Males Females Length of follow-up (mo)
(s)
Eighty-one patients with a first attack of isolated optic neuritis, 40 with Uhthoff’s symptom (Group 1)and 41 without (Group 21, were studied. All had a neurovisual examination, 74 of 81 patients had the pattern visual evoked potential recorded at rest, and 43 had magnetic resonance imaging brain scans. The pattern visual evoked potential PlOO latency was prolonged, Group 1 with a mean of 136 f 19 msec, Group 2 with a mean of 131 +- 19 msec (control subjects, 102 5 msec; n = 84), and the PlOO amplitude was reduced, without significant difference between the groups. Abnormal magnetic resonance imaging scans were present in significantly more patients in Group 1 ( p < 0.025). Treatment of optic neuritis with corticosteroids had no effect on the evolution or duration of Uhthoff’s symptom. Overall, 35 of 81 (43%) patients, followed for a mean of 3.5 years, developed multiple sclerosis. The incidence was significantly greater in Group 1 (p < 0.01). Uhthoff‘s symptom also correlated with a higher incidence of recurrent optic neuritis. We conclude that Uhthoff‘s symptom is a prognostic indicator for the early development of multiple sclerosis.
*
Scholl GB, Song H-S, Wray SH. Uhthoff’s symptom in optic neuritis: relationship to magnetic resonance imaging and development of multiple sclerosis. Ann Neurol 1991;30:180- 184
Uhthoff s symptom of intermittent transient blurring of vision on exertion has been described in multiple sclerosis (MS), isolated idiopathic optic neuritis, and in other optic neuropathies El-51. The symptom may also be provoked by emotional stress, temperature change, menstruation, increased illumination, eating, drinkmg, and smoking [G, 71. An inverse Uhthoffs symptom of improved vision has also been reported 18-10]. The pathophysiology of Uhthoff s symptom is unknown, although a reversible conduction block in demyelinated nerve fibers secondary to an increase in body temperature {11-141, or to changes in blood electrolyte levels or p H [l5, 161, is believed to play a role. The present study was designed to determine the incidence of Uhthoff s symptom in patients with a first attack of isolated optic neuritis and to compare the visual signs, the results of magnetic resonance imaging (MRI) of the brain, and the pattern visual evoked potentials (PVEP) of patients with and without Uhthoff s symptom. We chose to look at the PVEP at rest to see whether the neuropathy was accompanied by a delay in the Pl00 wave of equal magnitude in patiems with and without Uhthoffs symptom and if the range of
prolonged latency fell into the range consistent with a demyelinating process. The frequency of an abnormal MRI scan characterized by multifocal white matter lesions, in patients with clinically isolated optic neuritis, has been reported between 62 and 66% 117-131, but to date no study has compared the incidence of abnormal MRI scans in patients with and without Uhthoffs symptom. By looking at the incidence of an abnormal MRI scan, in conjunction with long-term follow-up for the development of clinical MS, we aimed to determine if the presence of Uhthoffs symptom could be predictive for the conversion of isolated idiopathic optic neuritis to MS.
From the Unit for Neurovisual Disorders, Department of Neurology, Massachusetts General Hospital, Boston, MA.
Address correspondence to Dr Scholl, Unit for Neurovisual Disorders, ACC 83?, Massachusetts General Hospital, 15 Parkman Street, MA 02’14’
Received Dec 12, 1990, and in revised form Feb 11, 1991. Accepted
Patients and Methods
Patients Eighty-one unselected patients with a first attack of acute idiopathic optic neuritis between February 1980 and December 1988 were studied in the Unit for Neurovisual Disorders at Massachusetts General Hospital (Boston, MA). None had a prior diagnosis of MS nor symptoms suggestive of demyelination. All patients were asked if they experienced temporary changes in vision with recovery back to baseline and, if so, what factors provoked the episodes. Seventy-five patients
for publication Feb 19, 1991.
180 Copyright 0 1991 by the American Neurological Association
had unilateral optic neuritis and 6 bilateral. There were 50 females and 3 1 males, aged 9 to 5 5 years; the mean age of females was 30 years and the mean age of males was 31.5 years. The patients were divided into the following two groups: Group 1, patients with Uhthoffs symptom, and Group 2, patients without Uhthoffs symptom. In Group 1, there were 40 patients, that is, 17 males and 23 females, aged 15 to 50 years, with a mean age of 30.9 years; in Group 2, there were 41 patients, that is, 14 males and 27 females, aged 9 to 55 years, with a mean age of 30.2 years. The PVEP was recorded in 84 normal control subjects with no history of eye disease, Uhthoffs symptom, or significant systemic disease, and a normal neuro-ophthalmic examination. There were 35 males aged 10 to 58 years, with a mean age of 30.7 years, and 49 females aged 8 to 74 years, with a mean age of 36.9 years.
Methods All patients had a neuro-ophthalmic examination and a workup to establish the diagnosis of acute idiopathic optic neuritis. Snellen visual acuity (VA) was recorded and expressed as a decimal (20/20 equivalent to 1.0). Follow-up permitted documentation of the worst VA and the best final VA. Fifty-eight patients, 33 in Group 1 and 25 in Group 2, also had a neurological examination to detect signs of MS. Additional studies to help establish a diagnosis of MS included somatosensory evoked responses, brainstem auditory evoked responses, and an examination of the cerebrospinal fluid (CSF). The clinical criteria of Perkin and Rose [20] for the diagnosis of MS in patients who have had optic neuritis were used for the clinical classification of MS into definite, probable, or possible. The additional category of laboratorysupported definite MS (criteria of Poser and coworkers 121)) was included for those patients with isolated optic neuritis and an abnormal CSF consistent with demyelination. Patients with an abnormal MRI scan but no other evidence of MS apart from optic neuritis were assigned to a fifth category of MRI-supported MS (criteria of Paty and colleagues 1191). The incidence of MS was calculated separately using these three diagnostic systems. A second attack of optic neuritis was not used in this study as a criterion for the diagnosis of MS. Patiems were followed for a mean of 3.5 years (Group 1 mean, 38 months; Group 2 mean, 45 months). Noncontrast MRI scans of the brain were obtained in 43 patients, that is, 23 in Group 1 and 20 in Group 2. 1'1 and T2-weighted images in the axial, sagittal, and coronal plains were obtained. Scans were interpreted as strongly suggestive of MS if there were three or more white matter lesions of high intensity on T2-weighted images, each measuring greater than 3 mm in diameter 1191. The PVEP was recorded in 35 of 40 patients in Group 1 and in 39 of 41 in Group 2 using methods reported elsewhere 122). All recordings were made during or after recovery from the initial attack of optic neuritis and before any subsequent attack. The latency and amplitude of the first major negative (N70) and positive (P100) peaks were measured. Check size 50' was chosen as the most likely to detect a difference in latency or amplitude in the two groups of patients. Responses are more frequently absent to smaller check sizes in patients with optic neuritis and visual loss.
Statistical analysis of the data was performed to look for significant correlations between the presence of Uhthoff s symptom and the other variables studied. Probability values were calculated by Fisher's exact test, x2 test, or Spearman correlation as appropriate and as indicated in the results.
Results Uhthoffs symptom was experienced by 40 of 81 (49.5%) patients with idiopathic optic neuritis at some time in the course of the attack and by 13 of 8 1 (16%) patients within 2 weeks of the onset of optic neuritis. In 1 patient, it commenced 6 years after the onset of optic neuritis, but most commonly, the symptom began within 2 months of the onset (23 of 40 patients). In 3 patients with early onset Uhthoff s symptom, there was a long period free of the symptom ( 3 , 14, and 32 months) followed by return of the symptom without a recurrent attack of optic neuritis. Episodes often continued over several months o r years. The frequency of episodes varied from sporadic to several episodes per day. Some patients noted an increase in frequency in the summer months. The provoking factors of Uhthoff s symptom are listed in Table 1. Exertion was the most common (21 of 40 patients), then hot baths or showers (11 of 40 patients) and hot weather (11 of 40 patients). T h e typical attack described by most of our patients was blurring of vision in the affected eye, which was usually generalized but was sometimes more pronounced centrally, coming o n after approximately 5 to 20 minutes of exposure to the provoking factor. A few patients noted color desaturation. Typically, vision recovered to its previous level over a variable period after resting or moving away from heat, commonly within 5 to 60 minutes, but sometimes taking u p to 24 hours. An athlete reported bilateral blurring of vision after 15 minutes of running, which after resting would remain blurred for 2 0 minutes, then recover to normal over the next 10 minutes. Only 1 patient experienced an inverse Uhthoffs symptom with improvement in vision under conditions of extreme stress at administrative meetings. The sensitivity to provocation of
Table 1 Factor5 Provoking Uhthaffs Symptom I
Factor
Patients (n)
Percentage
Physical exertion Hot bath or shower Hot weather Stress, anxiety, anger Tired, end of day Hot food or drink Cooking Other specific activity"
21 11 11 5 4
52.5 27.5 27.5 12.5 10 7.5
3 2 3
5 7.5
"Working a cash register, playing the trumpet, and reading (1 patient each).
Scholl et al: Uhthoff's Symptom
181
Table 2. Sarnmty of Clinical Data
Patients (n) Worst acuity at onset (mean) Worst acuity 50.3 (%) Best recovered visual acuity (mean) Recovered to 1.0 Treated with steroids (%) Having single attack (%) Having multiple attacks (%) Incidence of MS (%) Clinical criteria" Clinical and CSFb Clinical, CSF, and MRI' Incidence of MS' by sex (%) Males Females Length of follow-up (mo)
(s)
