Introduction

Isolated case reports of symptomatic cobalt intoxication after broken ceramic–metal and metal–metal hip joint endoprosthesis have been described since the early 2000s [1,2,3,4]. Typical diseases of cobalt intoxication are cardiomyopathy, hypothyroidism and neurological impairments with polyneuropathy, visual and hearing loss [5,6,7]. The reference value of the blood cobalt concentration is 0.45 μg/L [8]. To date, only one case has been published with severe cobalt intoxication above 600 μg/L, also following a hip prosthesis fracture [9]. Metallic abrasion of damaged prosthetic material causes the release of cobalt ions, which accumulate in the tissue and can lead to multiple toxic organ damage via systemic distribution in the blood. Little is known about the toxicological excretion kinetics of cobalt and chromium and their clinical course over several years. Additionally, there is little experience with potential chelation therapies such as EDTA (ethylene diamine tetraacetic acid) or BAL/DMPS (dimercaprol/2,3 dimercapto-1-propanesulfonic acid), whose efficacy has been judged unclear even in the presence of good renal function [1]. What is unique about this case report is the description of the long-term follow-up of clinical symptoms associated with the in vivo toxicology of cobalt and chromium in urine and plasma without chelation therapy.

Patient Information

In March 2018, a 63-year-old Caucasian male patient (BMI 27.7 kg/m2) with fatigue, decreased vision and hearing, metallic taste, difficulty falling asleep, nocturnal urination, hand tingling paraesthesia and numbness in the feet for 3 years presented to our medical facility for further diagnosis and treatment. Myocarditis and hyperacusis had already been diagnosed three years earlier, as well as hypothyroidism 2 years earlier and optic nerve atrophy in the last year. Because of these serious health limitations, he had to give up his business. Prior to this condition, the patient had not been diagnosed with any chronic disease.

Due to coxarthrosis on the right side, a total hip arthroplasty (H-TEP) with ceramic-ceramic bearing couple (KER/KER-H-TEP) had been performed in 2011. Because of a broken ceramic head, revision surgery of the right hip was accomplished in 2012 with a metal head/polyethylene (PE) bearing couple. Due to the broken hip joint prosthesis with chronic pain symptoms, another revision surgery followed 1 month before presentation with a change to a ceramic-ceramic bearing couple, this time a Delta ceramic (in 2011 it was still Forte ceramic) (Fig. 1, Table 1).

Fig. 1
figure 1

OP site in February 2018 with black staining of the tissue due to cobalt and chromium abrasion

Full size image
Table 1 Timeline
Full size table

Clinical Findings

At the initial presentation, 1 month after revision surgery, the patient showed reduced general condition and physical performance as well as signs of heart failure with nocturia. He reported central and peripheral nervous system dysfunction with numbness in the balls of his feet, slowed speech and slowed movement.

Diagnostic Assessment

Laboratory analysis revealed hypochromic microcytic anaemia (haemoglobin (Hb) 11.5 g/dl (mean corpuscular haemoglobin (MCH) 25.3 pg, mean corpuscular volume (MCV) 75.4 fl), normal values for white blood count, liver and kidney parameters, electrolyte status, total protein and TSH. Further toxicological laboratory results showed a blood concentration of 600 μg/L cobalt (reference 0.45 μg/L), 240 μg/L chrome (reference 0.4 μg/L) and 23.9 μg/L molybdenum (reference 16 μg/L).

Audiometry initially revealed bilateral hearing loss of 50 to 70 decibels (db(A)), particularly for frequencies of 1.5 to 3 kHz. The visual acuity test showed reduced capacity of the right eye: cc 0.05 and of the left eye: cc 0.25. Echocardiography showed reduced systolic left ventricular function (LV-EF 40%), dilated left ventricle (LV-EDD 64 mm), echo-rich myocardium, left atrial dilatation, mild mitral regurgitation and an initially small, hemodynamically irrelevant pericardial effusion (Fig. 2). At follow-up, left ventricular function had improved continuously to normal contractility (Fig. 3), as had hearing loss, numbness in the extremities and difficulty falling asleep. Psychiatric symptoms, weight loss, fatigue or dermatitis, which are also described in connection with cobalt intoxication, did not occur at the beginning or during the course.

Fig. 2
figure 2

Cardiac echo in 03/2018 with echo-rich myocardium and dilated left ventricle (Image courtesy Dr Paul Meyer)

Full size image
Fig. 3
figure 3

Echocardiographic follow-up of left ventricular ejection fraction (LV-EF—normal value: > 55%) over time line

Full size image

Therapeutic Intervention, Follow-up and Outcomes

During the 3-year follow-up after revision surgery in 2018, general condition improved and blood count normalized. At the last follow-up in 2021, the metallic taste in the mouth, hearing loss, visual loss and hypothyroidism requiring substitution persisted. After the long period of steady deterioration, the patient and his relatives were very pleased with the significantly improved quality of life due to normalized physical capacity and regained cerebral performance. In addition, a new-onset absolute arrhythmia was diagnosed in 2021 with atrial fibrillation. However, the patient had also suffered from arterial hypertension for years, so the cause of this new-onset arrhythmia does not appear to be clearly determinable at present.

During the entire follow-up period, possible drug therapy with EDTA or BAL/DMPS was not used as the cobalt and chromium ion concentrations had already decreased significantly four weeks after elimination of the source of poisoning. Close monitoring of the patient and the pathological toxic components was performed from 2018 to 2021, as shown in Figs. 4 and 5.

Fig. 4
figure 4

Blood concentrations of cobalt and chromium 2018–2021, respective reference values in parentheses

Full size image
Fig. 5
figure 5

Urine concentration of cobalt and chrome 2018–2021, also related to creatinine excretion. The respective reference values are shown in parentheses

Full size image

Discussion

Cobalt intoxications are often caused by inhalation or ingestion. Since 2003, a few cases of “prosthetic hip-associated cobalt toxicity” (PHACT) due to cobalt resorption from damaged endoprostheses have been described [10]. In a review of 10 representative individual case reports of cobalt intoxication from damaged hip prostheses, it was shown that the average plasma cobalt level of patients was 352 μg/L (minimum 23 μg/L, maximum 625 μg/L) and symptoms occurred on average 19 months (minimum 3 months, maximum 72 months) after cobalt exposure [11]. A PHACT with the highest known cobalt concentration was published in 2009; Oldenburg et al. described a cobalt concentration of 625 μg/L and similar symptoms as in our case [9]. In a study of 50 asymptomatic hip joint prostheses, wearers with metal prostheses mean cobalt plasma concentrations were found to be within the reference range with 1.0 μg/L at 1 year and 0.7 μg/L at 5 years [12]. Nordberg and Costa (2022) also described cobalt concentrations measured in the synovial fluid of 0.4 to 6 μg/L in asymptomatic patients who had undergone hip revision [13].

Consultation with the endoprosthesis manufacturer about the metallurgical composition of the prostheses used in our case revealed that, according to the ISO 5832-XII standard, they consisted of the following proportionate alloys: > 60% cobalt, 26–30% chromium (III), 5–7% molybdenum, < 1% nickel, manganese, silicon, iron, carbon and nitrogen. The toxic LD50 value for cobalt salts ranges from 150 to 500 mg/kg [14]. The chromium (III) compound is considered harmless to health (in contrast to chromium (VI) compounds, which, however, do not occur in endoprostheses) [15].

In the present case, cobalt and chromium were released from the metal femoral head after the first revision surgery. The ceramic particles remaining in the surgical site acted like a rubbing file on the newly inserted metal prosthesis. The clinical symptoms described here reflect the typical cardiac, thyroid and neurological effects of cobalt intoxication [16, 17]. However, these symptoms were not initially associated with the revised hip arthroplasty and possible intoxication by the treating physicians. The causative mechanism of cobalt-induced cardiomyopathy is discussed to be an accumulation of cobalt ions in the myocardial tissue, leading to a hypoxemic effect at the carotid chemoreceptors. In our case, a strikingly echo-rich myocardium was observed on sonographic imaging. However, the pathological mechanism in cardiomyocytes has not yet been fully elucidated [18].

Our case with a plasma concentration of 600 μg/L associated with PHACT cobalt intoxication is among the highest concentrations published to date. Surgical revision resulted in a rapid reduction of plasmatic cobalt and chromium concentrations within a few months and, accordingly, in a relatively rapid improvement of symptoms; however, the reference values were not reached (Figs. 4 and 5). Therefore, we decided—after the revision of the broken hip joint—not to use additional chelation therapy. The good patient adherence to close follow-up was of great advantage in this watch-and-wait approach. In general, the administration of chelating agents such as DMPS (2,3 dimercapto-1-propanesulfonic acid) is a drug therapy option that can lead to a reduction of the cobalt level in the blood. However, this therapy is sometimes associated with considerable side effects (e.g. general plasmatic metal salt elimination, eczema, allergic asthma) [5]. Blood cobalt concentrations of > 5 μg/L are described by Leavelle (2001) as probably toxic, but there are no robust data on this, so the use of chelators should still be considered critically [19]. Animal models show a relatively short elimination half-life of 42 to 156 h after intravenously administered cobalt, but high tissue accumulation with a large volume of distribution (0.94 l/kg) has been described [20]. In the present case, it must be assumed that the persisting blood concentrations of cobalt and chromium above the reference range are fed by a continuous release from the tissues. The strength of our case report is that it provides a substantial understanding of the long-term elimination kinetics of cobalt and chromium (III). However, a confounding factor in the evaluation of elimination kinetics is the continued release of cobalt ions from the contaminated tissue.

Conclusion

When neurological, cardiac and thyroid symptoms occur in patients with previous endoprosthetic hip replacement, rapid investigation of cobalt-induced intoxication should be considered. Priority should be given to eliminating the causative agent, e.g. by revision surgery of broken metal endoprostheses. The additional application of chelating agents should be carefully weighed up from the point of view of the risk–benefit ratio. With intact renal function, a rapid decrease in cobalt concentration and symptoms can be expected even with high initial concentrations of cobalt in the blood. Close combined clinical and toxicological monitoring is necessary and allows a watch-and-wait approach.