Depending on which discipline is studying pica, it has been interpreted as an eating disorder, a behavioural problem, a symptom of mental illness, an expression of nutrient deficiency, abandonment, neglect, poverty or hunger and, from a cultural point of view, (not a disorder) as a “beneficial” adjustment; at times a cause and other times a consequence, at times a well-defined entity and other times a non-specific symptom of some underlying process. The complications of pica found in the general population can be exacerbated in patients with chronic kidney disease (CKD), as some of these problems already occur as a result of the kidney damage. For example, pica can cause severe anaemia, electrolyte imbalance and problems absorbing micro and macronutrients. It may also lead to exacerbation of malnutrition, although that and the worsening of ingestion remain unproven due to the lack of scientific articles on this subject.1

With this article we present the current evidence on the characteristics, frequency and consequences of pica in patients with CKD. To identify studies assessing the presence of pica and associated factors in patients with CKD, we carried out a search in the databases Medline, Embase, Cochrane Database, EBSCO, Psychology and Behavioral Sciences Collection and PsycINFO, with combinations of the following set of keywords: pica, eating disorder, rumination syndrome, chronic kidney disease, kidney damage, renal failure, nephropathy, dialysis, renal replacement therapy, epidemiology, prevalence, risk factors, consequences, complications. To include as much information as possible, there was no restriction on time frame or place of publication, as long as they were published in English or Spanish, as we did not have the support of translators from other languages. The results of our literature search are described throughout this article.

Pica is one of the most dangerous self-destructive behaviours of people with mental disorders. Unfortunately, in up to 30% of cases seen for this type of complication, the history of pica is unknown, meaning they had not been diagnosed prior to this episode. More importantly, we know that 37% of patients are repeat offenders, i.e. they are being treated for a new episode even though they were on treatment, and this highlights the great difficulty of completely eliminating this type of behaviour.2Table 1 shows a list of epidemiological studies3–9 in populations vulnerable to pica (pregnant women, children and subjects with autism) and their clinical and pathophysiological findings. Studies on pica and pregnancy show that most pregnant women with pica have low haemoglobin and haematocrit levels and iron and zinc deficiencies. In children with pica, however, more common findings are malnutrition, intestinal parasitic disease and anaemia. As we will discuss below, there are some similarities between these populations and patients with CKD, particularly in terms of having anaemia.

The term “pica” comes from the Latin word for the magpie, a bird belonging to the crow family whose scientific name is Pica pica.10

The Diagnostic and Statistical Manual of Mental Disorders [DSM], version five (DSM-V), identifies pica as a term used to define a disturbance of eating or eating-related behaviour, which is characterised by its relationship with emotional or mental disorders,11 for which pica is defined as persistent eating of a substance(s) or object(s) with no calories or nutritional properties for at least one month in the last year, when it is not associated with a cultural and/or community practice and when it is considered inappropriate for the person's stage of development. Furthermore, if the eating behaviour occurs in the context of another mental disorder (e.g. intellectual disability, autism spectrum disorder, schizophrenia) or medical condition (including pregnancy and CKD) it is serious enough to warrant additional clinical attention.11

Treating physicians who suspect pica in their patients or who identify associated risk factors should evaluate them with the DSM-V to verify the diagnosis of pica, identify the substance ingested and control the symptoms. Fig. 1 shows an algorithm for the diagnosis of pica. The physical examination is one of the most important investigations for identifying the type of pica and/or related complications and should include exploring for signs and symptoms such as hair and nail loss, broken teeth, gingivitis, mouth ulcers and/or foreign bodies in the mouth. Blood analysis can be useful to identify anaemia, nutrient deficiency, electrolyte imbalance and/or impaired liver function. Testing of stools can identify the presence of parasites and/or of insoluble silicates eliminated 24–48h after the last meal.12 In some cases, chest or abdomen X-rays may be necessary to identify foreign objects or coins, or the presence of earth and/or hair.13

Fig. 1.

Algorithm for the diagnosis of pica.

Adapted from: American Psychiatric Association,11 Woywodt and Kiss,12 Waller and Pendergrass.13

(0.47MB).

Even with all the above tests, in most cases the clinical findings are non-specific, and diagnosis of pica then requires the patient (or family member) to admit the persistent behaviour. As there may be stigma attached to the diagnosis, the interview should be directed by a healthcare professional, aiming to use a non-judgmental approach to questioning the patient, and so help them to admit their problem more readily.13 Moreover, patients will often not have typical signs and symptoms any different from the underlying disease with which the pica may be associated, so skill and clinical acuity are essential. The most common scenario is that the diagnosis is not made until pica-related complications arise.1

The identification and monitoring of patients with pica is currently inadequate and healthcare personnel are wholly unprepared in terms of knowledge and training to identify this type of disorder at an early stage. This results in the complications and consequences spiralling out of control, even within healthcare institutions.14

Although pica has been studied by anthropologists, geographers, paediatricians, gynaecologists, haematologists, psychiatrists, psychologists, nutritionists, etc., the reality is that nobody knows what causes it.12 Suggested causes include nutritional deficiencies and social and psychological situations. However, at least in some patients, one of the main pathophysiological explanations for pica is addictive behaviour (similar to that of nicotine), as the behaviours/cravings often continue long after the physiological cause is resolved.15 It has been demonstrated that the behaviour persists even after the psychological cause is treated and alleviated. Moreover, if pica was always caused by nutritional deficiencies, it should stop when they are corrected, but that is not usually the case.16

Stillman and Gonzalez17 point to pica as a manifestation of obsessive-compulsive disorder. They state that patients on dialysis who display pica behaviours can consume a large number of substances which are associated with the alteration in their metabolic functions, although this is not always reflected in abnormal laboratory values.

CKD can be a powerful factor for emotional stress, and may therefore stimulate the onset of pica, especially in patients with a cultural predisposition.18 Patients with pica have described their behaviour as ritualistic and compelling, and their consumption as providing relief from anxiety.19 For example, in the Cooksey et al. study20 the patients stated that eating ice helped them when they felt stressed.

In general, factors such as abandonment and lack of supervision of children and people with disabilities by parents, poverty, hunger, gastrointestinal discomfort (for example nausea and vomiting), increased saliva production, and altered taste and smell have all been implicated in the origin and perpetuation of pica.11,21

Unlike the causes of pica, the complications are well documented. Of the complications of pica which occur in pregnant women, children and people with autism, Maravilla and Berk22 identified four types which can be of particular concern from a nephrology point of view: (1) toxicity of the substances (for example, lead or other substance poisoning); (2) gastrointestinal obstruction (for example due to consumption of hair, earth and/or stones); (3) excessive calorie consumption (for example, starch craving); (4) calorie deprivation because the ingested substance contains no calories and also suppresses the appetite (for example, ice consumption).

One of the most remarkable complications is the formation of bezoars (masses of indigestible substances that cannot progress and are isolated or trapped in gastrointestinal cavities); one of the most serious of these is Rapunzel syndrome (mass of hair anchored in the stomach that projects to the small intestine), found in children and people with mental disorders or with trichotillomania and/or trichophagia.23

In view of the impact of pica-related behaviours on the health of patients with CKD, from electrolyte imbalance to poor intake of food and adherence to diet because they feel sated from eating other substances such as ice or earth, and prefer their substance of choice to the diet prescribed by their health practitioner, healthcare personnel have to be able to detect patients with pica and prevent the associated complications and risk of malnutrition.24

Treatment of pica has to address the causes and consequences. Table 2 shows a summary of the treatment25 according to the type of pica, along with complications and predisposing factors. Psychopharmacology experts believe that there is no specific drug for treating pica. Generally, the use of drugs is in accordance with the psychiatric illness, with behavioural disorders predisposed to treatment and/or they are part of a comprehensive approach that includes psychological, social and educational measures.

One common psychological treatment is the application of aversive stimuli (for example the aversion to taste technique, where a bitter substance is mixed with the material consumed by the patient and they then develop an aversion to it). However, this has to be complemented by other psychoeducational strategies which aim to involve the patient in the process and help them understand more about pica and the associated risks.25 Some cognitive/behavioural treatment strategies applied in other disorders can be adapted for the treatment of pica26; by identifying what is going through the patient's mind when they start eating the non-food substances, it can be possible to modify this behaviour. It is also important to identify emotions and behaviours patients have as consequences of their dysfunctional thoughts, in order to help them restructure the underlying cognitive schemas.25 The inclusion of nutritionists, family support groups, motivational interviewing, anxiety management techniques, training in social skills and assertiveness training, individual therapy when necessary and prevention of relapse are basic elements in the treatment of pica. A multidisciplinary intervention of this type tends to be more effective than psychopharmacotherapy as a sole treatment.11 In any event, there is no one single method of treatment. What is essential is the work of a professional team prepared to investigate and take into account biological, psychological and social factors, who also understand the magnitude of the problem, how to approach it and the short- and medium-term consequences. Also important is educating the family and carers to give them basic ideas about how to control pica stimuli.26

Few studies have analysed the epidemiology of pica in a population with CKD. The available evidence is only in patients on long-term dialysis, haemodialysis (HD) and peritoneal dialysis (PD), and most of the information in this area comes from case reports. Table 3 shows the main results of studies performed in patients on dialysis, in whom the prevalence of pica varies from 10% to 46%17,18,27–34: Stillman and Gonzalez17 reported 38% in their sample of 292 subjects with dialysis; Obialo30 reported 22% in a sample of 138 African-Americans on haemodialysis; Katsoufis29 found a prevalence of 46% in a paediatric population of 87 children; Ward and Kutner18 a prevalence of 16% in 226 incident patients on replacement therapy; Ojanen28 found that among 41 patients assessed, 20% had some type of pica; and, lastly, Litt et al.,27 in the oldest study found published in 1984, reported a rate of 10%. However, Litt et al.27 did not specify how the individuals were questioned, which calls into question the validity of the results. Most studies include population on HD, and the main form of pica is the consumption of ice. In the majority of cases, low haemoglobin and haematocrit levels were found, as well as electrolyte concentration abnormalities, although with no well-defined pattern.

CKD can be a powerful factor for emotional stress, and that may act as a trigger for pica, especially in patients with a cultural predisposition.30 Due to the high prevalence of malnutrition and biochemical abnormalities in the population with CKD and dialysis, we need to identify the magnitude of pica and analyse its association with sociodemographic, clinical, biochemical and dietary factors, and so create the opportunity to establish the following strategies: (1) determine whether or not it is necessary to screen for pica; (2) identify factors that may be associated with the prevalence of pica and thus establish the necessary strategies to correct them; (3) determine whether or not pica is associated with the nutritional status of the patient on dialysis.

Establishing the diagnosis of pica (persistent and compulsive consumption of objects that have no calorific value) in a patient can be particularly complex. The most common scenario is that the diagnosis is not made until pica-related complications arise.

The causes of pica have not been completely clarified; psychological and nutritional causes have been proposed.

The complications of pica are associated with the toxicity of the substances consumed, gastrointestinal obstructions and alterations in calorie intake, especially relevant in patients with CKD.

The presence of pica in the patient with CKD may have a greater impact on the complications of kidney damage, such as anaemia, electrolyte and nutrient imbalances and malnutrition.

The authors declare that they have no potential conflicts of interest related to the contents of this article.