Clinical manifestations

Presentation and course

Various expressions are used by the patients to describe taste and smell alterations (see Table 1). A complete loss of smell or taste is rare, and most drug-induced smell and taste disorders involve reduced or distorted sensation.

Table 1. Terms for Taste and Smell Dysfunction

The most common types of dysgeusia associated with drugs are hypogeusia and a "metallic taste" parageusia.

The suffix -osmia is derived from Greek οσμή (osmí), odor, whereas olfaction, the sense of smell, is όσφρηση (ósfrisi); therefore, in the late 19th century, some considered terms like anosmia and parosmia to be misnomers, preferring instead anosphresia and parosphresia (60). Although anosphresia and parosphresia are still occasionally used in an idiosyncratic or pedantic manner, anosmia and parosmia are widely accepted and preferred.

Landis used the term "euosmia" to refer to the perception of "'false' but pleasant odors" in response to an odorant, a "pleasant parosmia" (28). However, this is an improper derivation. The prefix "eu" indicates the typical healthy state, not a substitution of something pleasant for something neutral or unpleasant, as in a pleasant smell for a neutral or a different pleasant smell. Thus, euosmia and eugeusia refer to normal olfaction and gustation, respectively. The case that Landis and colleagues describe would be better described as having heterosmia.

Prognosis and complications

Many drug-induced disorders of smell and taste are dose-related and resolve after drug discontinuation (06). In some cases, the loss of smell or taste or their distortion may be irreversible. Early detection of smell and taste disorders and prompt discontinuation of offending medications may prevent irreversible damage. Dysosmia and dysgeusia may be associated with weight loss due to a decrease in appetite, which results from a loss of perception of the flavor of food (34).

Biological basis

Etiology and pathogenesis

Disorders of smell and taste have been reported with numerous drugs. Some chemosensory complaints are due to the sensory properties of the drug itself, such as aversive bitter and metallic tastes (eg, with the drug secreted in saliva), but most chemosensory side effects of drugs are due to alterations in transduction pathways, biochemical targets, enzymes, and transporters (42). Studies of chemosensory adverse effects of drugs in medicated older individuals found that taste and smell loss is greatest in those taking the largest number of prescription drugs (42).

An analysis of drug‐induced taste disorders for drugs listed in the national drug information database for Dutch pharmacists found that, of the 1645 drugs registered in the database, 282 (17%) were documented to be associated with “dysgeusia” and 61 (4%) with “hypogeusia” (39). Drug‐induced taste disorders were reported in all drug categories, but predominantly in the following three categories: “antineoplastic and immunomodulating agents,” “anti-infectives for systemic use,” and “nervous system” (39). In about 45%, xerostomia (“dry mouth”) coincided as an adverse effect with taste disorders.

In an analysis of drug-induced smell and taste disorders in the FDA Adverse Events Reporting System (FAERS) database from 2011 to 2021, 16,091 adverse drug reactions were reported with olfactory and gustatory dysfunction, of which 13,641 (85%) were associated with gustatory reactions and 2,450 (15%) were associated with olfactory reactions (09). Zinc products (370 reports) and fluticasone propionate (214 reports) were most often associated with reduced olfaction (hyposmia/anosmia) (09). Varenicline (a smoking cessation aid; 24 reports) and fluticasone propionate (a glucocorticoid used to treat asthma, inflammatory pruritic dermatoses, and nonallergic rhinitis; 23 reports) were most often associated with parosmia (09). Lenalidomide (resembles thalidomide and is used to treat lymphomas, multiple myeloma, and myelodysplastic syndromes; 490 reports) and sunitinib (a kinase inhibitor, used as an antineoplastic agent; 468 reports) were most often associated with gustatory dysfunction (09). Antineoplastic and immunomodulating medications were responsible for 22% of olfactory adverse drug reactions and 36% and gustatory adverse drug reactions (09); among drugs in this category, immunoglobulin drugs were those most often associated with olfactory and gustatory adverse drug reactions.

Preclinical studies. Preclinical studies for drug safety in animals usually fail to detect sensory disturbances such as drug-induced smell and taste disorders (18). The impact of drug candidates on taste is rarely evaluated in preclinical toxicology studies during the early stage of drug development (58). However, in experimental studies on mice, the anticancer drug sonidegib, a Hedgehog-pathway inhibitor, led to rapid loss of taste buds in both fungiform and circumvallate papillae (27). Taste buds were not restored in all fungiform papillae even with prolonged recovery for several months, and this finding can explain the partial recovery of dysgeusia after discontinuation of sonidegib.

Clinical studies. There are large interindividual differences in susceptibility to smell-and-taste-related adverse effects, and the various factors involved include sex, age, body mass, and genetic variations in taste sensitivity. Loss in taste perception is more frequent in older individuals and is exacerbated by some drugs, diseases, and exposure to toxic chemicals.

Selected drugs for which the frequency of chemosensory disorders is reported at 1% or higher are listed in Table 2.

Table 2. Drugs Reported to Induce Chemosensory Disorders with a Frequency of 1% or More

Pathogenesis of drug-induced disturbances of smell and taste. Receptor dysfunction, which impairs the initial step of the sensory process, causes the bulk of drug-related smell and taste dysfunction. A very small minority involve either the central nervous system or sensory neural transmission.

Drugs may affect the receptors directly or indirectly by producing vitamin and essential element (zinc and copper) deficiencies. Zinc metalloproteins (gustin and lumicarmines for taste; a gustin-like protein for smell) are critical for maintaining receptor integrity. Molecular events underlying these disturbances can involve alteration of the primary, secondary, or tertiary receptor protein structure. The zinc-chelating ability of some drugs is associated with the development of drug-induced taste disorders (17).

Chemotherapy. The causes of chemotherapy-induced chemosensory dysfunction are likely multifactorial, probably with conductive (eg, thrush), sensory (altered taste buds), and neural components (35). With gynecological cancer chemotherapy, approximately half of the cases reported subjective taste complaints (35). Electrogustometry revealed a tendency for the development of hypogeusia in the chorda tympani nerve field, whereas hypergeusia tended to develop gradually in the greater petrosal nerve field (35). The filter paper disc test revealed a tendency for the development of hypergeusia for sweetness, saltiness, and sourness in the chorda tympani and glossopharyngeal nerve fields, whereas in the greater petrosal nerve field, there was a tendency for the development of hypergeusia for sweetness, sourness, and bitterness (35). Hypogeusia for bitterness tended to develop with an increasing number of chemotherapy cycles (35).

Chemotherapy changes the gene expression of T1R3 and T2R5 in head and neck cancer patients with mild/moderate stomatitis, resulting in dysgeusia and phantogeusia (55).

Marked taste disturbances are reported in cancer patients treated with Hedgehog-pathway inhibitor drugs, including sonidegib, which block the Hedgehog-pathway effector Smoothened (SMO) (27). Sonidegib treatment leads to rapid loss of taste buds in both fungiform and circumvallate papillae, including disruption of taste bud progenitor-cell proliferation and differentiation, but sparing non-taste papillae (27). In mice experiments, treatment cessation led to rapid and complete restoration of taste responses within 14 days associated with morphologic recovery in about 55% of taste buds, although taste buds were not restored in all fungiform papillae even after several months (27).

Dermatologic medications. An analysis of the most common dermatologic medications associated with smell/taste disturbances was reported in the Food and Drug Administration Adverse Event Reporting Database (59). About 60% of the drugs caused both olfactory and gustatory disturbances. The top 10 dermatologic medications associated with smell disturbances were (in descending order): Adalimumab (injection); Etanercept (injection); Terbinafine hydrochloride (oral); Cetirizine hydrochloride (oral); Vismodegib (oral); Secukinumab (injection); Prednisolone (oral); Spironolactone (oral); Isotretinoin (oral); and Fluconazole (oral). None had smell disturbances identified in clinical trials. The top 10 dermatologic medications associated with taste disturbances were (in descending order): Vismodegib (oral); Etanercept (injection); Terbinafine hydrochloride (oral); Apremilast (oral); Methotrexate (oral); Secukinumab (injection); Spironolactone (oral); Isotretinoin (oral); Tacrolimus (topical or oral); and Mycophenolate sodium (oral). With vismodegib, taste disturbances were reported in 67% of patients in phase III clinical trials, and smell disturbances were reported in a phase IV trial. With terbinafine, taste disturbances were reported in 3% of patients in phase III clinical trials, and taste or smell changes were described in 17 patients from an Italian adverse event reporting database. There are published case reports of chemosensory changes with (in descending order of the number of published reports) the following: Terbinafine hydrochloride (oral), n=9; Isotretinoin (oral), n=3; Vismodegib (oral), n=2; Etanercept (injection), n=2; Apremilast (oral), n=1; and Methotrexate (oral), n=1.

Antibiotics. An analysis of drug‐induced smell and taste disorders for antibiotics used in Japan identified significantly increased risks with azithromycin, clarithromycin, doxycycline, levofloxacin, and moxifloxacin hydrochloride (24). Fluoroquinolone and tetracycline antibiotics are known to inhibit zinc absorption (eg, by forming chelates with zinc). In addition, macrolide antibiotics decrease blood levels of divalent cations (calcium and magnesium). Based on the interactions among these metal cations and antibiotics, taste cell turnover and the olfactory epithelium may be adversely affected.

Antibiotics can potentially disturb gustation either through a primary effect of the drug or a secondary effect due to antibiotic-induced disturbances in oropharyngeal microbial flora. Primary effects may simply be due to secretion of a bitter-tasting drug (eg, clindamycin) into the saliva (10), or an antibiotic may inhibit or otherwise disrupt normal tastant/odorant receptor function (20).

Zinc-induced anosmia syndrome. Zinc-induced anosmia syndrome occurs after the exposure of olfactory epithelium to zinc contained in intranasal zinc gluconate gel used as a homeopathic over-the-counter "remedy" for the common cold. Commercial preparations of intranasal zinc gluconate gel were marketed even though intranasal zinc had been reported as a cause of anosmia (23; 02; Davidson 2006; Seidman 2006; 49; 08); patients diagnosed with zinc-induced anosmia or hyposmia reported sniffing deeply when applying the gel, followed by an immediate sensation of burning lasting minutes to hours, and subsequent loss of sense of smell within 48 hours (02; 49). The resulting severe hyposmia or anosmia appears to be long-lasting or permanent in some cases (23). Studies in animals (mice, rats, dogs) confirmed that zinc gluconate could cause anosmia and revealed that multiple divalent cations (eg, copper) could negatively affect olfaction (47; 21; 32; 12), whereas some found only transient impairment with large doses (48) or mild reductions in olfactory abilities despite severe zinc-induced reduction of afferent projections to the olfactory bulb (46).

Drug-induced burning mouth syndrome has been reported with antiretrovirals, antiseizure drugs, hormones, and antihypertensive medications that act on the angiotensin-renin system (41; 05; 52; 40; 49; 45; 37).

Other. Neuroleptic-induced parkinsonism is associated with olfactory deficits (26).

Antithyroid thionamides impair chemosensation by binding to the Bowman glands in the olfactory mucosa, where they can cause extensive lesions (03).

Possible mechanisms to explain the effect of some drugs on taste and smell are listed in Table 3.

Table 3. Pathomechanisms of Drug-Induced Disturbances of Smell and Taste

Epidemiology

The true incidence of drug-induced taste and smell disturbances is difficult to determine because of the infrequency of reporting. The estimated incidence of taste and smell alterations induced by drugs in clinical practice ranges between 2% and 5%. Of the 150,000 cases recorded in the pharmacovigilance centers in France, only 68 (0.05% of cases) reported olfactory complications related to drugs (36). The number of patients who experience a bad or bitter taste after sumatriptan is 5% (04). Old age and polypharmacy are risk factors for drug-induced taste disturbances. Evaluation of an Italian database of 52,166 spontaneous adverse drug reactions reported revealed 182 cases (approximately 0.3%) of drug-induced taste and smell alterations that were sometimes unexpected and often persistent complaints of patients during pharmacological treatments (56).

In a survey of cancer patients receiving chemotherapy, 55% complained of a gustatory disorder, and 19% complained of an olfactory disorder; occurrences of olfactory disorder were significantly greater in patients who had gustatory disorder than in patients who did not (51). Similarly, about half of the patients undergoing chemotherapy for gynecological cancer report altered gustation (35).

In an analysis of the drug-induced taste disorders in the Dutch Pharmacist’s database, 17% were listed as dysgeusia and 3.7% as hypogeusia (39). They occur in all drug categories but predominantly in antineoplastic and immunomodulating agents, antiinfectives for systemic use, and those for neurologic disorders.

Prevention

Drug-induced smell and taste disturbances can be minimized by avoiding the use of drugs known to induce such disturbances in patients with diseases that make them susceptible to smell and taste disorders. Good nutritional support with zinc supplementation may reduce the possibility of the onset of drug-induced smell and taste disorders. Good oral hygiene coupled with prevention of dry mouth may reduce the incidence of taste disturbances. Once a patient shows signs of such disturbance, early discontinuation of the offending drug may prevent complete loss or irreversible distortion of smell or taste.

Differential diagnosis

Confusing conditions

Drug-induced olfactory and taste disturbances should be differentiated from those associated with aging and several diseases. Age-related loss of smell, presbyosmia, may occur with normal aging or neurodegenerative disorders associated with aging. Taste can also diminish with aging, and loss of smell also contributes to taste impairment.

Drug-induced olfactory disorders are usually bilateral, whereas unilateral anosmia is usually unrelated to drug use. Smell and taste disorders frequently coexist, and patients with loss of smell also have taste sensation impairment. Perceived taste disturbance might be an interaction of various health factors such as illness and mental condition and is difficult to differentiate from drug-induced taste disturbances. Exposure to low levels of toxins in the air over long periods can lead to transient olfactory disorders or even anosmia.

Diagnostic workup

Confirmation of drug-induced chemosensory disorders can be provided in some cases by drug rechallenge (16; 53), although generally this is not necessary, and in some cases, it can lead to more severe and lasting impairment.

Physical examination and office testing. Physical examination should include nose, mouth, and throat, and neurological examination.

Office testing of smell can include well-standardized, commercially available tests (such as the University of Pennsylvania Smell Identification Test, or UPSIT) or crude approaches utilizing identification of a few readily available odorants (such as oil of wintergreen or oil of cloves) (14; 15; 13); more complicated odor identification and detection tests are also available but are rarely practical outside specialized diagnostic laboratories (13). The UPSIT is a forced-choice olfactory discrimination test that uses microencapsulated odorants in standardized “scratch ‘n’ sniff” booklets. UPSIT scores are standardized by gender and age and can be used to identify degrees of hyposmia and some malingerers. Irritant substances, such as ammonia, are sometimes employed when psychogenic or malingered anosmia is a consideration because such substances are perceived via trigeminal afferent pathways instead of through the olfactory system.

Office testing of gustatory function remains somewhat crude (43). Testing of taste thresholds is problematic because salivary function and size of the tongue area stimulated influence threshold assessment, taste intensity may be depressed even with normal recognition thresholds, and changes in threshold detection do not necessarily correlate with suprathreshold taste intensity. Nevertheless, complicated and time-consuming procedures are available for whole-mouth assessment of thresholds for each of the four primary tastes (sweet, sour, salty, and bitter) (62; 63), but are also rarely practical outside specialized diagnostic laboratories. Newer technologies, such as using flavor-impregnated taste strips or tasting tablets, are in development and can potentially include assessment at multiple thresholds for all of the primary tastes, including umami (savory) gustatory sensation (01; 28).

Special examination methods for assessing gustation. Chemical testing uses chemical tastants (eg, sucrose for sweet, citric acid for sour, sodium chloride for salty, caffeine for bitter, and monosodium glutamate [MSG] for umami) to assess taste. Unfortunately, routine office testing of taste may not reveal any findings in transient, drug-induced taste disturbances. More complicated methods that can be used in office environments have been developed, but they are more typically used in subspecialty clinics or research settings devoted to chemosensory disorders.

The 3-drop method of chemical testing is a subjective, non-localizing ("whole mouth") assessment technique that uses aqueous solutions of the five5 basic tastes to provide both the detection threshold (ie, the concentration at which the subject correctly identifies the presence of a taste) and the recognition threshold (ie, the concentration at which the subject correctly identifies the particular basic taste) (43; 61; 38). For each forced-choice trial of three separate drops (the two control drops being just water), the subject is asked to select the one that contains the tastant. Larger volumes can be used, in which case the subject must sip, swill, and spit. The tastant must be correctly identified three times, or the next higher concentration solution is used. One potential confounding factor with this approach is that the aqueous solutions may wash away the saliva, the medium through which tastants are normally conveyed to taste buds, thus, altering taste sensitivity (31). The use of tablets or wafers impregnated with the basic tastants eliminates some of the logistical challenges of using aqueous solutions of escalating concentration with controls; these can be easily stored and used for routine clinical practice but suffer the same drawbacks as any whole-mouth testing method (01; 43; 38).

Whole-mouth chemical testing methods are generally sufficient for drug-induced gustatory impairment, but, when necessary, regional chemical testing is possible with commercially available "Taste Strips": paper strips impregnated with one of four tastants (sweet, sour, salty, or bitter) in four different concentrations—16 separate strip types for all tastant-concentration combinations (33; 28; 43; 38). The Taste Strips are placed in a predetermined order on each side of the tongue while it is protruding from the mouth. Lateralizing hypogeusia is identified with side-to-side differences above an established threshold. Regional testing can detect isolated lesions of the chorda tympani (anterior two-thirds of the tongue) or glossopharyngeal nerve (posterior tongue).

Electrogustometry is a subjective, localizing assessment technique that uses a small electrical current (1.5 to 400 μA) from a monopolar or bipolar stimulator to assess "taste" detection thresholds in different parts of the tongue. This test can provide localizing information about gustatory deficits (43), but it cannot distinguish between the different qualities of flavor because electrical stimulation invariably elicits a metallic and sour "taste" sensation. Approximately half of the women who receive chemotherapy for gynecological cancer have abnormal electrogustometry test results, with a tendency for the development of hypogeusia in the chorda tympani nerve field and hypergeusia in the greater petrosal nerve field (35): both pathways are extensions of the facial nerve, but the chorda tympani originates from the taste buds in the front of the tongue, whereas the greater petrosal nerve (or greater superficial petrosal nerve) is a branch of the nervus intermedius (nerve of Wrisberg or glossopalatine nerve) that, inter alia, carries taste afferents from the soft palate.

Gustatory evoked potentials are a feasible, objective assessment technique that can distinguish ageusia from psychogenic disease or malingering (25; 22; 38). After application of a chemical or electrical taste stimulus, evoked waveforms are recorded from an EEG. Unfortunately, various confounding factors that may influence gustatory evoked potentials (eg, thermal, tactile, or olfactory stimulation) are difficult to interpret and only provide consistent results with sour tastes (38).

Imaging cranial imaging (CT and MRI) is most useful for patients with olfactory disorders to detect abnormalities in the nasal cavities (eg, nasal polyposis), nasal sinuses (eg, sinusitis), and anterior cranial fossa (eg, meningioma).

Management

Treatment is mostly limited to zinc supplementation when zinc deficiency is confirmed. Treatment can also include dose reduction or discontinuation of the offending drug or the substitution of the offending drug by another in the same therapeutic category. If the drug-induced disturbances are severe enough to impair the quality of life and the drug is not required for a serious or life-threatening condition, it should be discontinued.

In a double-blinded, placebo-controlled, randomized clinical trial on patients with chemotherapy-induced alterations in taste or smell, zinc was ineffective (30).

Patients receiving platinum-containing chemotherapy for ovarian cancer may suffer olfactory and gustatory impairment, which usually resolves 3 months after completion of therapy. Patients should be informed of this side effect, and symptomatic relief can be provided by additional flavoring of food and a small amount of glutamate (50).

Cyclophosphamide causes drug-induced losses of taste sensory cells and disruptions of proliferating cells that renew taste sensory cells, resulting in dysgeusia and hypogeusia. Pretreatment with amifostine can protect the taste system from adverse effects of cyclophosphamide (11); amifostine protects taste cell renewal and the population of cells within taste buds from the cytotoxic effects of cyclophosphamide with few observable adverse effects due to repeated administration. Amifostine is already approved to reduce the nephrotoxic effects caused by repeated treatment with cisplatin for advanced ovarian cancer and xerostomia caused by radiation therapy after surgery in some patients with head and neck cancer.

In patients with chemotherapy-induced chemosensory dysfunction, supplementation with lactoferrin, a globular glycoprotein, decreases the concentration of salivary iron and produces an overall increase of expression of immune proteins, whereas it ameliorates the adverse effects on taste and smell (57). Other treatment options to ameliorate smell and taste disturbances include the addition of simulated flavors to food to compensate for losses and to override offending tastes and smells (42).

Outcomes

Early treatment by withdrawing the offending drug provides good chances for recovery of smell and taste. Late cases with damage to the olfactory nerve may not recover.

Special considerations

Pregnancy

Pregnancy has not been reported to affect drug-induced smell or taste disturbances.

Anesthesia

Smell and taste disturbances may result from nerve injury during infiltration with local anesthetics used during procedures on the teeth or the nose. These may occur as effects of general anesthetics as well.