In 1997, Brown et al1 reported on theuse of substance P and insulin-like growth factor I for the topical treatmentof neurotrophic and anhidrotic keratitis in a child. Since then, other authorshave used these compounds2,3 totreat a variety of corneal conditions that caused chronic focal or diffuseepithelial disruption. Herein, we report a possible complication when thecombination of substance P and insulinlike growth factor I was used to treatneurotrophic keratitis in a child with complete trigeminal nerve (cranialnerve V) palsy.
A 28-month-old girl sustained permanent, total left trigeminal nervepalsy after a motor vehicle crash. She was referred to us 3 months later witha chronically red eye and partial ptosis. The mother reported that the childwould vigorously rub and pick at her left nostril (causing bleeding), scratchher cheek, and also dig her fingers into her left inferior conjunctival fornix,without evidence of pain. On examination, the inferior conjunctiva was diffuselyinjected. The exposed inferior third of the cornea was rough and partly opaque,with adherent mucous strands. The cornea underneath the ptotic lid was clearand lustrous. Partial oculomotor nerve palsy prevented an adequate Bell reflex.Lacrimation could not be tested, but an adequate, although mucoid, tear lakewas present. Formal testing of corneal and periocular sensation was not attempted.The diagnosis was traumatic, neurotrophic, and exposure keratopathy. Initialtreatment with heavy topical lubrication and erythromycin ointment failedto prevent frequent recurrent erosions and superimposed bacterial keratitis.A lateral tarsorrhaphy was performed, but the cornea failed to stabilize,partly because of the child’s self-mutilating behavior.
On the basis of our previous experience, we elected to treat the childwith substance P (Multiple Peptide Systems, Inc, San Diego, Calif), 250 μg/mL,combined with insulinlike growth factor I (Boehringer-Mannheim GmbH, Mannheim,Germany), 25 ng/mL, dissolved in hyaluronate sodium (Healon GV; Pharmacia& Upjohn, Inc, Kalamazoo, Mich), 1 drop 3 times daily. Investigationaldrug approval was obtained from the US Food and Drug Administration. The firstscheduled follow-up visit at 1 week was not completed. The child was seen2 weeks after initiating treatment. Her mother reported that after 2 to 3days, she developed white blisters along the lower eyelid margin, and rednessand blistering of the lower lid and the cheek. The mother discontinued themedication after 7 days of treatment. The lid margin blisters resolved rapidly.
Seven days after discontinuation, the inferior conjunctiva was somewhatless injected than before treatment. The chronic inferior corneal erosionhad partly reepithelialized at its nasal and temporal corners. The lid marginswere normal. The Figure shows the leftlower eyelid and the skin of the cheek.
Photograph taken 7 days after discontinuationof substance P–insulinlike growth factor I combination treatment showshyperemia of the lower lid and an ulcerated skin lesion. The lid margin blisteringreported by the mother is no longer evident.
To the best of our knowledge, this is the first reported case in whicha combination of substance P and insulinlike growth factor I was used to treatneurotrophic keratitis in a patient who also had trigeminal trophic syndrome.Trigeminal trophic syndrome occurs in patients with facial paresthesias dueto trigeminal nerve injury.4 As in this child,compulsive nose picking is a typical feature. Our patient had a mild manifestationof this syndrome, which often can result in complete loss of the nasal aladue to persistent self-mutilation. After several days of topical ophthalmicadministration of the substance P–insulinlike growth factor I combination,the mother observed a blistering or vesicular reaction involving the lowereyelid margin and surrounding skin, which began to resolve when the medicationwas discontinued. We have treated 2 previous children with a combination ofsubstance P and insulinlike growth factor I, both with congenital cornealanesthesia alone, who did not develop a cutaneous reaction during many weeksof therapy.
There are several possible explanations for the vesicular eruption seenin our patient. First, a substance P agonist has been shown to mediate contacthypersensitivity in the skin.5 Both irritantcontact dermatitis and allergic contact dermatitis have been shown to be enhancedby neuropeptides, including substance P.6 Contacthypersensitivity may have occurred when up-regulated cutaneous substance Preceptors were suddenly exposed to normal or supranormal levels of substanceP. Irritant contact dermatitis and allergic contact dermatitis can be seenas vesicles in the acute stages of the eruption. Second, the vesicular eruptioncould have been a herpetic infection (probably due to herpes simplex virus1). The latter may have occurred coincidentally or may have been stimulatedby the substance P–insulin-like growth factor I combination therapy.No previous causal relationship between substance P–insulin-like growthfactor I combination therapy and herpetic infections has been reported. Becausethe initial clinical impression was that this was a resolving dermatitis,viral cultures from the cheek ulcer were not obtained. Finally, we used thehyaluronate as the drug vehicle to increase surface retention and decreasethe dosing frequency to 3 times per day. In the previously reported case,we used balanced salt solution with a dosing frequency of 1 drop every 15minutes for 2 hours, twice a day. It is remotely possible that this case representsan allergic contact dermatitis to the hyaluronate vehicle. Rechallenge withthe substance P–insulinlike growth factor I topical therapy has notbeen performed.
Substance P combined with insulinlike growth factor I has proved effectivein several cases of chronic neurotrophic keratitis unresponsive to customarytherapies.1,2 Patients withtrigeminal trophic syndrome, atopic dermatitis, or other types of allergicskin disease may have a cutaneous blistering reaction to these agents whenapplied to the surface of the eye. Increased contact hypersensitivity dueto substance P is one possible mechanism. Physicians should be alert to thispossibility when substance P (with or without insulin-like growth factor I)is used to treat recalcitrant neurotrophic keratitis in patients who haveperiocular cutaneous denervation.
Correspondence: Dr Brown, Department ofOphthalmology and Visual Sciences, Texas Tech University Health Sciences Center,3601 Fourth St, STOP 7217, Lubbock, TX 79430-7217 (email@example.com).
Financial Disclosure: None.
Brown SM, Khanani AM, Stetson CL, Reid TW. Vesicular Eruption in a Child With Trigeminal Nerve Palsy During TopicalTherapy With Substance P and Insulinlike Growth Factor I for NeurotrophicKeratitis. Arch Ophthalmol. 2004;122(11):1722-1723. doi:10.1001/archopht.122.11.1722