Clinical Problems in the Hospitalized Parkinson's Disease Patient: Systematic Review

The problems Parkinson's disease (PD) patients encounter when admitted to a hospital, are known to be numerous and serious. These problems have been inventoried through a systematic review of literature on reasons for emergency and hospital admissions in PD patients, problems encountered during hospitalization, and possible solutions for the encountered problems using the Pubmed database. PD patients are hospitalized in frequencies ranging from 7 to 28% per year. PD/parkinsonism patients are approximately one and a half times more frequently and generally 2 to 14 days longer hospitalized than non-PD patients. Acute events occurring during hospitalization were mainly urinary infection, confusion, and pressure ulcers. Medication errors were also frequent adverse events. During and after surgery PD patients had an increased incidence of infections, confusion, falls, and decubitus, and 31% of patients was dissatisfied in the way their PD was managed. There are only two studies on medication continuation during surgery and one analyzing the effect of an early postoperative neurologic consultation, and numerous case reports, and opinionated views and reviews including other substitutes for dopaminergic medication intraoperatively. In conclusion, most studies were retrospective on small numbers of patients. The major clinical problems are injuries, infections, poor control of PD, and complications of PD treatment. There are many (un-researched) proposals for improvement. A substantial number of PD patients' admissions might be prevented. There should be guidelines concerning the hospitalized PD patients, with accent on early neurological consultation and team work between different specialities, and incorporating nonoral dopaminergic replacement therapy when necessary. © 2011 Movement Disorder Society


Introduction
Parkinson's disease (PD) is the second most common neurodegenerative disorder with a life-time risk of 2 percent in men and 1.3 percent in women. 1 Although the disorder is generally slowly progressive, it does have a major impact on disability and quality of life of affected patients. 2,3 One of the lesser studied aspects of PD is the spectrum of problems PD patients encounter, once they are admitted to a hospital. In our own and others' experience hospital admissions of PD patients are often problematic, especially so when patients are admitted on non-neurological wards. 4,5 Problem areas are exact timing or lack of drug administration, administration of contra-indicated drugs, complications due to immobilization, and psychiatric disorders triggered by the hospital admission. [6][7][8] As most non-neurologically educated health care personnel are unfamiliar with PD, protocols would be helpful to improve the care of PD patients in such environments.
We recently surveyed the majority of movement disorder specialist neurologists in the Netherlands and found that no specific guidelines or protocols exist to guide caregivers and PD patients in the hospital environment (manuscript in preparation). Before such guidelines can be formulated, one needs to know the prevalence and spectrum of the problems PD patients may experience during their hospital stay. To this end, we systemically reviewed all the existing literature on the problems encountered by the hospitalized PD patients.

Methods
We systematically reviewed the literature on reasons for emergency room (ER) and hospital admissions in patients with PD, problems encountered during hospitalization of this patient population, and possible solutions for the encountered problems, using the Pubmed database. Last research date was 17 June 2010.
To identify articles, we included (combinations of) keywords: See Table 1 for search details. Subsequently we analyzed the abstracts for relevant articles: as relevant articles we defined those as pertaining to the following four areas: 1. Analysis of prevalence and reasons for ER visits and subsequent admission. 2. Clinical problems during hospital stay. 3. Peri-and postoperative problems. 4. Suggestions for improvement of care for the hospitalized PD patient.
We also searched the reference list of each relevant article for other applicable articles.
In our search, there were no language limitations. We excluded articles concerning brain surgery.
As most studies vary greatly in selection of patients, exact relative proportions of these 3 groups cannot be assessed (Tables 2 and 3). Some studies found that PD patients are more likely to be admitted to the ER and hospital for complications of the disease and its management than for primary motor problems. 13,18 A part (5-21%) of the patients were first diagnosed to have PD during a hospitalization. 19,21 Problems During Hospitalization We found one prospective study on acute events occurring during hospitalization. 4 When admitted to a neurology ward (83 PD patients, mean age 69 years, mean disease duration 6 years), patients received an average of 0.6 non-neurological consultations. Reasons were: (aspiration) pneumonia, urinary infections and retention, diarrhoea, atrial fibrillation, postural hypotension, low back pain, and TIA. Specialists consulted most frequently were cardiologists, internal medicine specialists and orthopaedic surgeons. Acute events observed during hospitalization on non-neurological department of 20 patients (mean age 80 years) after ER visits were: Urinary infection (33%), agitation, confusion and hyperthermia (28%), pressure ulcers and leg oedema (22%), bowel occlusion and hypotension (11%), and dysphagia (6%). Mortality rate during hospitalization of this last group was 20%. We found one retrospective study on this subject (173 PD patients), but this study is unclear on the presence or absence of complications before the hospital admission. 21 Problems identified in this study do accord however with those described above. 4 Another retrospective study shows that in 74% of 35 emergency hospital admissions of PD patients, their medication was stopped, omitted or prescribed inappropriately with 61% of this group suffering significant sequelae. Nonadherence to medication schedules was a large problem and in 11% of cases antidopaminergic medication (metaclopramide) was prescribed. 31 In a study analyzing inpatient falls, the use of anti-Parkinson's medication was an evident risk factor (OR 5.04). 32 Apart from the studies above, we found several opinionated views and reviews on problems during hospitalization of PD patients. These authors point to disease fluctuations, stress, 7 and medications like butyrophenones, phenothiazines, and metoclopramide, prescribed during hospitalization, as possible causes of PD exacerbation. 6,[33][34][35][36][37][38] Additionally, sleep disorders, verbal communication problems, nutritional intake difficulties, and cognitive changes are mentioned as problem areas. 7,19,39

Perioperative Problems
We found 15 retrospective studies on problems during and after surgery in patients with PD, with an emphasis on the postoperative period. Of these 15, two retrospective studies included a control group of non-PD patients. 25, 27 One retrospective study focused on postoperative confusion 33 and two retrospective studies on medication errors. 40,41 Another study with 10 patients is actually a case report with 10 PD patients, 42 and the remaining 9 papers are retrospective studies on PD patients undergoing orthopaedic surgery. 29,[43][44][45][46][47][48][49][50] The first study compared 234 PD patients with a control group of 40,979, undergoing elective bowel   In the second study with a non-PD control group, 51 PD patients, treated on different surgical departments, were compared using matched-pair analysis. There were significantly more postoperative falls in the PD group (18% vs. 4%), and a higher although not significantly increased number of urinary tract infections (33% vs. 24%), pneumonia (10% vs. 4%), and also wound infection, urinary retention, respiratory insufficiency, and postoperative confusion. PD patients were hospitalized for more days and stayed on the intensive care unit longer. 27 Another retrospective study showed that 60% of 25 PD patients with no pre-existent mental status abnormalities suffered postoperative confusion, some with hallucinations. The onset of the delirium was often delayed, 70% after 36 hours. In this study, there was no relationship between delirium and type of anti-Parkinson medication or anaesthetic procedure. 33 In a retrospective study on pharmacological management during 51 surgical admissions of PD patients or patients with parkinsonian syndromes treated with PD medication, 71% had missed doses of their medication. Overall, antidopaminergic medication was prescribed in 41% and administrated in 22% although not allowed. 47% (69% for non-day-cases admissions) had complications: neuropsyciatric 41%, falls 8%, and worsening of motor symptoms 5%. 40 A second study on pharmacological management found that 30% of 92 PD patients had medication administration problems, leading to an increase in postoperative con-fusion or worsening of PD: 84% to 36% in the wellmanaged group. Overall 31% of patients was dissatisfied in the way their PD was managed in the perioperative period. 41 In the smallest study (n ¼ 10), all patients had complications. This paper identified the same problems as mentioned above, and additionally found pressure sores as an important problem. 42 Next to the above studies, there are nine retrospective research articles on orthopaedic surgery and its complications. These studies (totalling 433 PD patients, Table 4) found pneumonia, urinary tract infections, confusion, and decubitus as the most frequent postoperative complications, in frequencies up to 49% [44][45][46][47][48][49][50] with an overall six month mortality up to 47%. 44,45,47,48,50,51 Apart from complications, several studies are contradictory regarding the outcome of orthopaedic procedures in PD patients, 26,29,30,43-60 and postoperative rehabilitation is reported to be slower. 28,55 Next to these more or less formal surveys, case report and reviews mention the following perioperative problems: neuroleptic malignant syndrome, medication-or anaesthetic-induced exacerbation of PD, side-effects of PD medication, postextubation laryngeal spasm, bronchospasm, respiratory arrest, difficulty with salivation, gastrointestinal complications, deep vein thrombosis, urinary disturbances, temperature regulation problems, and tremor hampering eye surgery. 6,8,[34][35][36][37][38][61][62][63][64][65][66][67][68][69][70][71][72][73][74][75][76][77][78][79][80] Improvement of Care for the Hospitalized PD Patient

Improvement During Hospitalization
There are no studies analyzing the effects of suggested recommendations/improvements. Some authors favour a multidisciplinary approach during hospitalization. 4,7 Other suggestions are: continuing the exact

Improvement of Perioperative Care
There are only two studies on medication continuation during surgery 81,82 and one study analyzing the effect of an early postoperative neurologic consultation, 29 and numerous case reports, and opinionated views and reviews.
During preoperative screening, some authors recommend extra attention to be paid to this group of patients, especially to respiratory status, urologic system, fluid status, cardiovascular system, gastrointestinal system, autonomic system, and cognition. 6,25,35,37,66,73,74,78,83 And, if necessary, supple-mented with additional diagnostics like laboratory tests, pulmonary tests, electrocardiogram, and Xray. 25, 66,73,78 Most of the literature describes measures with regard to PD medication in the intraoperative period. To prevent large descents of dopamine levels intra-and postoperatively many authors advise continuation of PD medication as long as possible preoperatively and resume it as soon as possible postoperatively, and PD medication is preferably continued. 6,8,25,34,37,38,42,[67][68][69][70][72][73][74]76,78,79,81,82,[84][85][86][87][88][89][90][91][92] Two studies on this topic describe PD patients using the rotigotine transdermal patch, a nonergot D1/D2/D3 dopamine agonist. 81,82 In the first prospective study, oral dopaminergic medication was easily switched to rotigotine before surgery and resumed afterwards in 14 PD patients undergoing surgery under general anaesthesia. Adverse events were two dopaminergic side effects namely nausea and hallucinations and one ventricular asystole. 81 The second study on this topic describes PD patients derived from two prospective clinical trails. PD patients undergoing surgery under general anaesthesia and who continued using rotigotine during surgery were retrospectively analyzed (n ¼ 25). There was no worsening of PD symptoms, but (only) three complications: deep vein thrombosis, infection, and pain. 82 As other substitutes for dopaminergic medication intraoperatively, continuous intravenous levodopa infusion, continuous subcutaneous infusion or immediate postoperative injections of apomorphine, and enteral levodopa/carbidopa via nasogastric tube or duodenostomy have been used by various authors, 68,69,72,76,78,88,92,93 but none have been studied in a controlled trial. The use of parenteral anticholinergic and antihistaminic medication as anti-Parkinson therapy is limited according to some authors, and may aggravate postoperative discomfort because of autonomic side effects and confusion. 72,76,78 Some authors claim that general anaesthesia should be avoided when possible, and prefer local anaesthesia. 37,66,67,73,75,90 This is supported by a case report of a PD patient, undergoing surgery with regional anaesthesia, who was successfully given levodopa and carbidopa orally during the operation. 86 Regional anaesthesia also avoids postoperative nausea and vomiting. 73 Less invasive interventions, like laparoscopic surgery over abdominal surgery, are recommended. 66 Some authors advocate carefully considering the used operation technique. 47,50 In a retrospective study, the postoperative period was analyzed in PD patients undergoing total knee arthroplasty receiving a preoperative or immediate postoperative neurologic consultation (n ¼ 13) compared with patients receiving a delayed or no visit (n ¼ 21). Only in the first group, there was a significant improvement in total Unified Parkinson's Disease Rating Scale with most improvement in activities of daily living but also an improvement of mood, mentation and behavior, and motor examination. In this group, there was also a shorter length of stay (P < 0.01) and less patients were confused: 15 to 48%. 29 There are many more recommendations, without formal studies. Postoperatively lung expansion manoeuvres, noninvasive mechanical ventilation, percussion, adequate pain control, aspiration precautions, (early) mobilization, recognition of unique physical limitations and medication combinations, physiotherapy, turning regimens, maintenance of volume status, antiparkinsonian therapy adjustments, analyzing urine for urinary tract infections, and deep venous thrombosis prophylaxis may prevent complications like pulmonary emboli, infections, deep vein thrombosis and decubitus. 6,27,[42][43][44][45]50,55,59,60,66,72,74,78,83,91,[94][95][96] In case of prolonged endotracheal intubation early tracheostomy is preferred. 66 Except for two case reports, there is no literature concerning the pre-emptive administration of drugs to prevent complications. In one case report, the cholinesterase inhibitor rivastigmine was given preoperatively to prevent delirium postoperatively in a PD patient successfully treated with this drug in two previous delirious episodes. 97 In another case report, prokinetics were administered in two patients to prevent paralytic ileus. 92 Some authors suggest preventive antibiotics to prevent infections. 44,45,59 As described before, the onset of the postoperative delirium is often delayed. If patients are discharged rapidly after surgery, there should be sufficient support in the home environment. 33 Overall, team work between different specializations, like surgery, neurology, geriatrics, and a rehabilitation unit is generally advocated. 28,96,98 Discussion There are few studies analyzing the problems a PD patient encounters during hospitalization, and there are even less studies analyzing possible solutions. Most studies are retrospective and have small numbers of patients. In some studies, PD patients were diagnosed according to clear diagnostic criteria (like the United Kingdom Parkinson's Disease Society Brain Bank's clinical diagnostic criteria), 4,[9][10][11]15,19,48 but in most studies, the diagnostic criteria were not clearly mentioned: a neurologist confirmed the diagnosis, 13,43,45,46,48 or medical record systems and/ or patients notes were used to identify PD patients, 17,[20][21][22]25,27,33,40 or no information according to the diagnostic criteria, except the use of PD medication, was given. 16,31,41,42,47,49,50,81,82 Overall PD patients are more frequently 20,23,24 and longer 12,16,17,19,21,[24][25][26][27][28][29] hospitalized compared with controls. Generally the leading causes for admission are injuries (many with fractures), infections [mainly pneumonia and urinary tract infections (UTI)], poor control of PD and complications of PD treatment, psychiatric disturbances, and diseases of the circulatory and digestive system. A reduction of the number of admissions might be achieved by extra attention to fall prevention, adequate drug regulation with acuity for side effects, preventing and recognizing of early symptoms of infections and active monitoring in the home situation of both patients' vital parameters and therapy compliance. 5,18,20 When admitted to a hospital, most PD patients stay on a general medicine or surgery ward instead of a neurological one. 5,15 Apart from one small prospective study 4 and a retrospective study, 31 little is known about problems occurring during hospitalization of PD patients not undergoing surgery, mentioning the usual direct and indirect PD related problems and medication issues.
There are more studies on complications in PD patients hospitalized for surgery, particularly in the postoperative period. (Aspiration) pneumonia, UTI, bacterial infections, postoperative falls, postoperative delirium/confusion (often with a delayed onset 33 ), and hypotension occur more frequently in this group of patients than in controls. 25,27 Pressure sores are also an important complication. 42 Mortality rates are also higher. 25 These data are confirmed by studies on PD patients undergoing orthopaedic surgery. 29 administration problems with more postoperative confusion or worsening of PD as a result. 40,41 It is striking that almost one third of the PD patients were dissatisfied concerning their PD treatment. 41 Suggestions for Improvement of Hospital Care There are many proposals for improvement during hospitalization in PD patients with or without surgery. Suggestions on improvement do vary, but most authors agree that attention should be given to all aspects of PD and not only to motor function.
Most publications refer to the intraoperative period mainly to prevent a decline in dopamine levels because of discontinuation of dopaminergic medication, including two studies favouring the use of a rotigotine patch. 81,82 Most authors agree that anesthesiologists and surgeons should take the increased vulnerability of PD patients into account when planning and selecting procedures. 37,47,50,66,67,73,75,86,90 One small study shows that early consultation by a neurologist may prevent complications and reduce length of hospital stay. 29

Conclusions
Most studies were retrospective and had small numbers. Prospective studies with large numbers of PD patients, defined according to clear diagnostic criteria and preferable diagnosed by a specialist with special interest in PD, would be preferred for future research. Generally patients with PD are hospitalized much more frequently and longer than control groups. The leading causes are injuries, infections, poor control of PD and complications of PD treatment. The inclusion of hospitalization data into patient registries for PD could be a major improvement in identifying the important problems.
There are many (un-researched) proposals for improvement during hospitalization in PD patients. A substantial number of PD patients' admissions might be prevented. There should be guidelines concerning the hospitalized PD patients, with accent on early neurological consultation or consultation of another specialist like a geriatrician with a special interest in PD and team work between different specialities, and on sufficient training of all people involved in the treatment and recovery of this patient group. This protocol should include dopaminergic replacement therapy in case PD patients are not allowed or able to take their oral medication. Preferably this therapy should not be laborious and not invasive so that it is easily applicable.