Abstract
Background:
Cisplatin is a commonly used chemotherapy agent known to induce serious adverse reactions that may require hospital readmission. We aimed to analyze the extent and factors associated with unplanned hospital admissions due to cisplatin-based chemotherapy regimen-induced adverse reactions.
Methods:
Retrospective review of medical records of those patients who received at least one cycle of chemotherapy with cisplatin-based regimen during a six-month period from March to August 2017.
Results:
Of the 458 patients who received cisplatin during the study period, 142 patients did not meet inclusion criteria. The remaining 316 patients had a total of 770 episodes of primary admissions for chemotherapy administration. Overall, 187 episodes (24%) of intercycle unplanned hospital admission were recorded of which a major proportion (n=178; 23%) was due to chemotherapy-induced adverse reactions. Underweight patients had higher odds of unplanned admission (OR 1.77, 95% confidence interval [CI] 1.11 to 1.77). Significantly, more number of patients with cancers of head and neck and cancers of musculoskeletal were readmitted (p<0.001). Compared to high-dose cisplatin, low- and intermediate-dose cisplatin had lesser odds of unplanned admission (OR 0.52 and 0.77; 95% CI, 0.31 to 0.88 and 0.41 to 1.45, respectively). Patients without concomitant radiotherapy, drug-drug interaction and initial chemotherapy cycles had lesser odds of unplanned admission (OR 0.38, 0.50 and 0.52; 95% CI, 0.26 to 0.55, 0.25 to 0.99 and 0.32 to 0.84 respectively). Unplanned admissions were mainly due to blood-related (31%) and gastrointestinal (19%) adverse reactions. Among chemotherapy regimens, cisplatin monotherapy (34%) and cisplatin with doxorubicin (20%) regimens resulted in a major proportion of unplanned admissions.
Conclusion:
These findings highlight risk factors that help identify high-risk patients and suggest that therapy modifications may reduce hospital readmissions due to cisplatin-based chemotherapy-induced adverse reactions.
Keywords: Adverse drug reaction, hospital readmission, cisplatin, associated factors, cancer treatment, causality
1. INTRODUCTION
Cisplatin is a platinum derivative introduced several decades ago as an antineoplastic agent to treat a myriad of solid tumors. Though the oldest member of its group, it is still a key component of many chemotherapy regimens including those for bladder, head and neck, lung, ovarian, testicular cancers and musculoskeletal tumours [1]. However, among the widely used anticancer drugs, cisplatin is most commonly associated with various forms of adverse reactions. The drug is highly emetogenic, neurotoxic, nephrotoxic, and ototoxic [2]. The incidence rate of cisplatin-induced adverse reactions is as high as more than 90%[2]. Majority of these adverse reactions are usually mild and are self-limiting [3]. Nevertheless, an unknown proportion of these reactions are severe enough to require hospital readmission [4]. Hospital readmissions have been reported in up to 52% of patients who were on cisplatin-based chemotherapy regimen [5, 6]. However, these reports were based on a few specific cisplatin regimens and cancer types rather than cumulative cisplatin-based chemotherapy regimens.
Hospital readmissions are of particular concern because of its significant impact on cost and patient outcomes. Several studies have tried to explore factors that influence hospital readmissions among the general population. Poor care coordination, poor follow-up care, age, female gender, patients of lower socio-economic status, and shorter hospital stay are all known to elevate the risk for readmission [7, 8]. Regarding the causes of readmission, a study [9] analyzed diagnoses of potentially avoidable hospital readmissions. It found that in a majority of cases, the readmission diagnoses is a direct or indirect complication of patients’ primary diagnoses itself. Interestingly, the same study has found adverse reactions to drugs as one of the most frequent readmission diagnoses. Furthermore, patients with a cancer diagnosis have been reported to have an increased risk of potentially avoidable readmissions compared to other comorbidities [9].
Thus far, a majority of pertinent studies have only focused on hospital readmissions among general medical patients. However, existing evidence suggests that cancer patients, in particular, are more vulnerable to readmission [10]. Among cancer patients, patient characteristics including primary diagnoses, gender, comorbidities, number of previous admissions are all shown to be associated with readmission [10], where most of these factors are non-modifiable. Though reported, many studies have either overlooked or underestimated chemotherapy-induced adverse reactions as a cause of hospital readmission. In our study setting, we observed that among cancer patients, especially those who received cisplatin-based regimen, readmission diagnoses were often associated with adverse reactions. Since adverse reactions are potentially modifiable, readmissions due to adverse reactions could possibly be reduced by knowing the factors associated with it and by introducing appropriate interventions [11]. Generally, to prevent hospital readmissions, healthcare settings are now turning to predictive models in order to identify patients at high risk for readmission and focus resource intensive readmission prevention strategies on such “high-risk” patients [12].
Knowledge of which patients by age, gender, cancer type, cisplatin dose, number of chemotherapy cycle, concomitant drugs or chemotherapy, and drug-drug interactions have the greatest odds of readmission due to adverse reactions is important. This knowledge could help identify high-risk patients and direct the assessment and development of new opportunities to focus on readmissions. Therefore, the aim of this study was to estimate the rate and to investigate those factors associated with intercycle unplanned hospital admissions or readmissions due to cisplatin-based chemotherapy regimen-induced adverse reactions.
2. MATERIALS AND METHODS
This was a retrospective review of medical records of those patients who were diagnosed with any solid tumour and received at least one cycle of chemotherapy with cisplatin monotherapy or cisplatin-based chemotherapy regimen. The study was conducted in a 2300 bedded large tertiary care teaching hospital with established oncology and radiotherapy units. Institutional Review Board approval was obtained prior to initiation of this study.
Patients’ medication histories were obtained from the hospital’s pharmacy workstation that documents medication transactions of all in-patients and out-patients. Corresponding clinical data were obtained from the electronic medical records containing all hospital admission and readmission data. Unique hospital identity numbers were used to track an individual across pharmacy workstation and electronic medical records.
2.1. Study Population
We identified all patients who received at least one dose of cisplatin over a period of six months between March 2017 and August 2017 from the pharmacy workstation. We excluded outpatients and patients whose clinical and/or pharmacy data were either missing or incomplete. As the intent of the study was to identify various clinical and patient factors associated with hospital readmissions, we included special populations too (for example children, patients with multiple comorbidities and so on), which are typically excluded in other studies.
2.2. Outcome Measures
The primary outcome measure was intercycle unplanned hospital admission due to adverse reactions induced by cisplatin or cisplatin-based regimen. We defined intercycle hospital admission as hospital readmission that occurred between two consecutive chemotherapy cycles during the study period. We intentionally did not use the term ‘30-dayhospital readmissions’ which is widely used in literature for public reporting and comparing hospital readmissions because chemotherapy cycles, in particular, cisplatin regimens, are often given more frequently than 30 days interval [13]. Unplanned admissions for conditions other than those due to adverse reactions including those due to disease exacerbations were grouped separately and not accounted for analysis. Readmissions were attributed to adverse reactions only after conducting causality analysis using WHO-UMC system for standardized case causality assessment [14]. The readmission diagnosis and patient complaints were linked with the chemotherapy regimen administered in the previously planned hospital admission and current medications. Only those readmissions due to adverse reactions that were deemed certain, probable, or possible in the WHO-UMC scale were considered as the primary outcome. More than one unplanned admission within 30 days of discharge was counted separately.
Covariates included patient characteristics, cancer type, cisplatin dose, cisplatin regimen, concomitant drugs or radiotherapy, chemotherapy cycle, drug-drug interactions, and comorbidities. Age wise, patients were grouped into three categories (children ≤18 years; adults <60 years; elderly ≥60 years). According to body mass index (BMI), patients were categorized into underweight (<18.5 kg/m2), normal weight (18.5 to 24.99 kg/m2), and overweight/obese (≥25 kg/m2). Comorbidities were identified and classified into two (patients with no known comorbidities and patients with one or more comorbidities). Cisplatin dose was categorized into low (<60 mg/m2), intermediate (60 to 99 mg/m2), and high (≥100 mg/m2) [15]. Clinically significant drug-drug interactions were also identified after consulting Lexicomp Drug Interaction online tool and Stockley’s Drug Interactions (11th edition) and grouped into two including those without drug-drug interactions and those with one or more drug-drug interactions.
2.3. Statistical Analysis
Continuous variables were presented as the mean ± Standard Deviation (SD) or as a median and range, and these were compared using the Student t test or the Mann-Whitney U test as appropriate. Categorical variables were presented with a number and percentage, and these were compared using the Chi-squared test. We performed logistic regression to analyze the relationship between unplanned hospital admission and variables including age, gender, Body Mass Index (BMI), cisplatin dose, chemotherapy cycle, concomitant radiotherapy, concomitant drugs, drug-drug interaction, comorbidities, and type of cancer. For clinical outcomes, the statistical significance was derived by adjusting the values for potential confounding factors such as patient age, BMI, and comorbidities. A p value of less than 0.05 was considered significant. All analyses were performed with SPSS v16.0.
3. RESULTS
Screening the pharmacy workstation yielded a total of 458 patients who received at least one cycle of cisplatin-containing chemotherapy regimen during the study period. Of these, 142 (31%) patients either received cisplatin in an outpatient/daycare setting or had incomplete information and hence were excluded. The remaining 316 patients resulted in a total of 770 episodes of primary admissions for chemotherapy administration resulting in an average admission rate of 2.44 episodes per patient during the study period. Patients were relatively younger with a mean age of 35.64 ± 17.69 and more likely to be male (66%). In a majority of the episodes, patients received their first chemotherapy cycle (41%) and received low dose cisplatin (47%). In 21% of episodes patients received concomitant cisplatin monotherapy with radiotherapy, while 16% of patients had one or more comorbidities. One or more clinically significant drug-drug interactions were observed in 17.1% of episodes.
The overall intercycle unplanned hospital admission rate was found to be 24% (n=187). After conducting causality analysis, nine readmissions (4.8%) were found due to reasons other than chemotherapy-induced adverse reactions and hence excluded from further analysis. Hence the actual rate of unplanned hospital admission due to a cisplatin-based chemotherapy-induced adverse reaction was 23% (n=178). A majority of these adverse reactions were associated with blood and lymphatic system disorders such as anaemia, neutropenia, thrombocytopenia, and pancytopenia. Patients readmitted due to gastrointestinal disorders developed diarrhoea, nausea, vomiting, gastroenteritis, abdominal pain, dysphagia, and/or anorexia. Metabolism and nutrition disorders included dyselectrolytaemia such as hyponatraemia, hypomagnesaemia, and hypokalaemia. General disorders included acute febrile illness, chills, pain, dizziness, and weakness. Patients readmitted with infection had sepsis or septic shock. Table 1 shows the full demographics and clinical characteristics of patients.
Table 1. Demographics and clinical characteristics of patients.
| Characteristics | n | % |
|---|---|---|
| Gender Men Women |
506 264 |
65.7 34.3 |
| 30-day hospital readmission status Readmission due to adverse reactions Readmission due to other reasons No readmission |
178 9 583 |
23.1 1.2 75.7 |
| Cisplatin dose Low-dose Intermediate-dose High-dose |
360 313 97 |
46.8 40.7 12.5 |
| Concomitant radiotherapy Yes No |
162 608 |
21.0 79.0 |
| Concomitant drugs Cisplatin monotherapy One concomitant drug Two concomitant drugs Three concomitant drugs |
176 335 251 8 |
22.9 43.5 32.6 1.0 |
| Drug-drug interaction status No interactions Had one interaction |
638 132 |
82.9 17.0 |
| Characteristics | n | % |
| Comorbidities No comorbidities One or more comorbidities |
647 123 |
84.0 16.0 |
| Chemotherapy cycle Cycle 1 Cycle 2 Cycle 3 Cycle 4 and above |
315 190 138 127 |
41.0 24.7 18.0 16.5 |
| Cancer types Genitourinary and gynecology Head and neck Musculoskeletal Thoracic and respiratory Brain and nervous system Gastrointestinal Endocrine Others |
250 158 114 92 83 46 18 9 |
32.5 20.5 14.8 11.9 10.8 6.0 2.3 1.2 |
| Adverse reactions Blood and lymphatic system disorder Gastrointestinal disorders Metabolism and nutrition disorders General disorders Infections and infestations Nervous system disorders Respiratory Renal Others* |
55 34 28 28 17 5 4 4 3 |
30.9 19.1 15.7 15.7 9.6 2.8 2.2 2.2 1.7 |
*Others included ‘immune’, ‘hepatobiliary’, and ‘skin’.
3.1. Patient Characteristics and Hospital Readmissions
Age and gender of patients were not found to have a significant association with unplanned hospital admissions. However, patients’ BMI was found to have an association (p=0.033). In multivariable analysis, underweight patients were associated with higher odds of readmission (OR 1.77, 95% confidence interval 1.11 to 1.77), whereas overweight and obese patients remained unaffected (p=0.741). The increase in the number of comorbidities was not found to be significantly associated with unplanned hospital admission due to adverse reactions.
Unplanned admission rates varied significantly (p<0.001) by certain types of a cancer diagnosis. Head and neck cancer and musculoskeletal tumours together constituted a significantly high proportion of patients who got readmitted (28.7% and 21.9%, respectively). However, these cancer types constituted only a small proportion of those who did not require readmission due to adverse effects (18.1% and 12.7% respectively) and the difference was found to be statistically significant. As given in Table 2, other types of cancer did not differ significantly between those who were readmitted and not readmitted.
Table 2. Characteristics of 30-day readmission and non-readmission of patients.
| Characteristics |
Readmission
n=178 (%) |
No Readmission
n=592 (%) |
P value |
|---|---|---|---|
| Gender Male Female |
113 (63.5) 65 (36.5) |
393 (66.4) 199 (33.6) |
0.474 |
| Age | 37.03 ± 17.10 | 35.23 ± 17.86 | 0.233 |
| Cisplatin dose Low dose Intermediate dose High dose |
89 (50.3) 55 (31.1) 33 (18.6) |
271 (45.8) 258 (43.6) 64 (10.6) |
0.002 |
| Concomitant radiotherapy No Yes |
116 (65.2) 62 (34.8) |
492 (83.1) 100 (16.9) |
<0.001 |
| Drug-drug interaction No interaction One or more interactions |
159 (89.3) 19 (10.7) |
479 (80.9) 113 (19.1) |
0.009 |
| Comorbidities No comorbidities One or more Comorbidities |
145 (81.5) 33 (18.5) |
502 (84.8) 90 (15.2) |
0.287 |
| Chemotherapy cycle Cycle 1 Cycle 2 Cycle 3 Cycle 4 and above |
84 (47.5) 31 (17.5) 36 (20.3) 27 (14.7) |
231 (39.0) 159 (26.9) 102 (17.2) 100 (16.9) |
0.040 |
| Concomitant drugs Cisplatin monotherapy One concomitant drug Two or more concomitant drugs |
60 (33.7) 65 (36.5) 53 (29.8) |
116 (19.6) 270 (45.6) 206 (34.8) |
<0.001 |
| Cancer Type Musculoskeletal Genitourinary and gynecology Head and neck Thoracic and respiratory Others* |
39 (21.9) 50 (28.1) 51 (28.7) 21 (11.8) 17 (9.6) |
75 (12.7) 200 (33.8) 107 (18.1) 71 (12.0) 139 (23.5) |
<0.001 |
*Others included bone and muscle, brain and nervous system, breast, endocrine system, gastrointestinal, haematopoietic and skin.
As shown in Table 3, after adjusting for patients’ clinical and demographic characteristics, we observed a significant variation in adverse reactions-related unplanned admissions across different cisplatin dosages (p=0.01). Referenced by patients who received high-dose cisplatin, the odds ratio for unplanned admission decreased by half for those who received low dose cisplatin (OR 0.52, 95% confidence interval 0.31 to 0.88). Relative to a low dose, the odds ratio for unplanned admission slightly increased for those who received intermediate dose cisplatin but remained below that of high-dose cisplatin (OR 0.77, 95% confidence interval 0.41 to 1.45).
Table 3. Factors associated with 30-day hospital readmission due to chemotherapy-induced adverse reactions.
| Variable |
Unadjusted
OR |
95%CI | P value |
Adjusted
OR |
95% CI | P value |
|---|---|---|---|---|---|---|
| Gender Male Female |
0.88 1.00 |
0.62-1.25 Reference |
0.475 | 1.27 1.00 |
0.80-2.01 Reference |
0.309 |
| Age ≤18 years 19 to 45 years >45 years |
1.00 1.04 1.13 |
Reference 0.66-1.62 0.72-1.77 |
0.881 0.605 |
1.00 1.02 1.16 |
Reference 0.57-1.83 0.61-2.18 |
0.936 0.655 |
| Variable |
Unadjusted
OR |
95%CI | P value |
Adjusted
OR |
95% CI | P value |
| BMI Underweight Overweight/Obese Normal |
1.77 0.91 1.00 |
1.77-1.11 0.91-0.53 |
0.016 0.741 |
NA | ||
| Cisplatin Dose Low dose Intermediate dose High dose |
0.63 0.41 1.00 |
0.39-1.02 0.24-0.68 Reference |
0.059 0.001 |
0.52 0.77 1.00 |
0.31-0.88 0.41-1.45 Reference |
0.014 0.415 |
| Radiotherapy No radiotherapy was given Radiotherapy was given |
0.38 1.00 |
0.26-0.55 Reference |
<0.001 | 0.54 1.00 |
1.71-16.90 Reference |
0.004 |
| Drug-drug interaction No Interaction (reference) One or more interactions |
1.97 1.00 |
1.18-3.31 Reference |
0.010 | 0.50 1.00 |
0.25-0.99 Reference |
0.048 |
| Comorbidities No Comorbidities Oneor more Comorbidities |
0.79 1.00 |
0.51-1.22 Reference |
0.287 | s | NA | |
| Chemotherapy cycle Cycle 1 Cycle 2 Cycle 3 Cycle 4 and above |
1.40 0.75 1.36 1.00 |
0.85-2.30 0.42-1.34 0.76-2.41 Reference |
0.187 0.329 0.297 |
0.52 0.90 0.69 1.00 |
0.32-0.84 0.55-1.47 0.40-1.18 Reference |
0.008 0.676 0.174 |
| Concomitant drugs No concomitant drugs One concomitant drug ≥2 concomitant drugs |
2.01 0.94 1.00 |
1.30-3.10 0.62-1.40 Reference |
0.002 0.748 |
2.67 4.32 1.00 |
0.80-9.09 1.26-14.80 Reference |
0.112 0.020 |
| Cancer type Musculoskeletal Genitourinary and gynecology Head and neck Others Thoracic and respiratory |
1.76 0.85 1.61 0.41 1.00 |
0.94-3.27 0.48-1.51 0.89-2.91 0.21-0.83 Reference |
0.075 0.568 0.113 0.013 |
0.22 0.89 0.61 0.26 1.00 |
0.11-0.47 0.40-1.99 0.27-1.37 0.12-0.55 Reference |
<0.001 0.776 0.233 <0.001 |
Adverse reactions-induced unplanned admissions were significantly higher among patients who received concomitant radiotherapy (p<0.001). Referenced by patients who received radiotherapy, the odds ratio for unplanned admission was significantly lower for those who did not receive radiotherapy (OR 0.38, 95% confidence interval 0.26 to 0.55).
The number of concomitant drugs had a significant effect on adverse reactions-induced unplanned admissions. The odds of unplanned admission among those who received cisplatin monotherapy was twice as that of those who received one or more concomitant drugs (OR 2.01, 95% confidence interval 1.30 to 3.10). However, this difference was not found to be statistically significant when adjusted for confounding factors. Drug-drug interactions within the chemotherapy regimen were also significantly associated with adverse reactions-induced unplanned admissions. In multivariable analysis, the absence of interaction was associated with the least odds of unplanned admission (OR 0.50, 95% confidence interval 0.25 to 0.99). Table 4 lists the chemotherapy regimens involving clinically significant drug-drug interactions. Similarly, the odds of unplanned admission steadily increased as the number of chemotherapy cycle increased (for example, OR 0.52, 95% confidence interval 0.32 to 0.84 for the first cycle vs the fourth cycle or above).
Table 4. List of chemotherapy regimens involving drug-drug interactions.
| S. No. | Chemotherapy Regimen | n | % |
|---|---|---|---|
| 1 | Paclitaxel and cisplatin | 53 | 40.2 |
| 2 | Docetaxel, cisplatin, and fluorouracil | 48 | 36.4 |
| 3 | Docetaxel and cisplatin | 10 | 7.5 |
| 4 | Paclitaxel, ifosfamide, and cisplatin | 8 | 6.0 |
| 5 | Cetuximab, docetaxel, and cisplatin | 8 | 1.5 |
| 6 | Bevacizumab, paclitaxel, and cisplatin | 5 | 4.5 |
Among the adverse reactions, 31% of unplanned admission episodes were due to blood-related reactions, whereas gastrointestinal disorders resulted in 19% of unplanned admissions. Other important adverse reactions resulting in unplanned admissions included metabolic disorders, general disorders and infections.
Among chemotherapy regimens, cisplatin monotherapy had induced adverse reactions and resulted in unplanned hospital admissions in 34% of episodes. Cisplatin with doxorubicin was the second most common chemotherapy regimen with an unplanned admission rate of 20%. Twenty other regimens contributed to rest of the unplanned admission episodes.
4. DISCUSSION
Cisplatin-induced adverse reactions are very common and are often serious and/orsevere. Cisplatin-based chemotherapy regimens further complicate the clinical stability of patients by exposing them to additional adverse reactions, rendering patients more vulnerable to adverse reactions and contributing to drug-drug interactions. Patients with cancer diagnoses are known to be a cohort with a higher risk of potentially avoidable readmissions, whereas adverse reactions were found to be frequent readmission diagnoses [9]. Cisplatin is one of the most widely used anticancer drugs with relatively greater potency for adverse reactions. The target population in this study included cancer patients who received cisplatin-based chemotherapy regimen during the study period. Moreover, in our clinical practice, we found that cancer patients, in particular, those who were on cisplatin-based chemotherapy regimens were highly susceptible to develop serious adverse reactions that often required hospital readmissions. In this study, we found that intercycle unplanned hospital admissions among patients who received cisplatin-based chemotherapy regimen were very common due to adverse reactions. Among patient characteristics, underweight patients were found more susceptible to adverse reaction-induced unplanned admissions. Other important factors emerging that had a stronger association with unplanned admission included the type of cancer diagnosis, high cisplatin dose, concomitant radiotherapy, drug-drug interactions and number of chemotherapy cycles.
Large retrospective studies have estimated the 30-day readmission rate in cancer patients between5% and 25% [16-18]. In addition, the readmission rate widely differs between types of cancer and different chemotherapy regimens. For example, previous studies have reported 22% readmission rate among patients with gastrointestinal cancers, 13.8% among patients with head and neck cancer, 4.3% among those who underwent lung cancer surgery and 52% in those who received etoposide, ifosfamide, and cisplatin [5, 19, 20]. To the best of our knowledge, ours is the first study of its kind in cancer patients who received cisplatin-based chemotherapy regimens. We found an all-cause unplanned admission rate of 24% in our cohort of patients, whereas the rate due solely to adverse reactions is 23% (95% of overall all-cause readmissions).
One study reported a hospital readmission rate of 31% due to cisplatin-induced adverse reactions [6]. But in this study, the cohort was limited to those with head and neck cancer receiving low-dose cisplatin concurrent with radiotherapy. Although the readmission rate is almost similar to what we have found, our study results are more generalizable with respect to age, cisplatin regimen, type of cancer and concurrent radiotherapy.
In contrast to the general population, age was not found to influence unplanned hospital admissions in our target population. A recent large retrospective analysis [21], which included patients with all conditions, explored a nonlinear association with age, in which the likelihood of readmission was elevated for children transitioning to adulthood. However, it is interesting to note that our results are similar to that reported in the literature when the cancer cohort alone is considered [22]. Neither gender was found significantly associated with unplanned admission due to adverse reactions in our study population, unlike Hispanic and Taiwanese cancer patients in whom male gender had a higher risk of all-cause readmission [22, 23]. This study underscores the importance of body weight in unplanned admissions. Previous studies have shown that overweight and obese patients are at higher risk for all-cause readmission in both adults and children with different conditions including heart failure [24-26]. However, in stroke patients the opposite has been reported in which readmission rate was significantly lower among overweight and obese patients, whereas significantly higher in underweight patients when compared to normal weight patients [27]. Our study results concur with those of stroke patients in which underweight patients had a higher risk, whereas overweight and obese patients were at a lesser risk for readmission compared with that of normal weight patients.
Our study results showed that half of those admitted patients had one or more comorbidities, but this factor was not found to have a significant association with adverse reaction-induced unplanned admission in our cohort. Manzano et.al.reported that patients with three or more comorbidities and or metastatic disease had a higher risk of readmission [28]. However, our cohort significantly varied with respect to the number of comorbidities. In our cohort, a majority of patients had either one or two comorbidities and rarely more than two. This variation might explain the difference in effect which needs further investigation to find the true effect of this factor.
Generally, patients with concomitant medications are perceived to have a higher risk of adverse reaction-induced readmissions. However, we observed a paradoxical effect of higher odds of readmissions in cisplatin monotherapy group when compared to those with one or more concomitant medications. The administration of concomitant radiotherapy might help to explain the increased odds of unplanned admission observed in the cisplatin monotherapy group. In the monotherapy group, patients received radiotherapy in 98% of episodes compared to that of 2.5% in the other group. Previous studies have shown that radiotherapy could possibly increase the severity of chemotherapy-induced adverse reactions [29].
Patients in our study were more likely to be readmitted due to blood-related adverse reactions followed by gastrointestinal disorders and electrolyte imbalance. We attributed the adverse reactions to cisplatin regimen by conducting a causality assessment using one of the commonly used tools, WHO-Uppsala Monitoring Centre tool. Previous studies using informal assessment methods have been known to report artificially inflated numbers of adverse reactions [30]. In addition, the pattern of adverse reactions in our study corroborates with those reported in the product and medical literature [31]. Though chemotherapy-induced adverse reactions are very common, they are poorly reported or documented in medical records. Specific comprehensive medical terminologies such as SNOMED-CT and MedRA are used to document adverse reactions by adverse event reporting systems including MedWatch and VigiFlow. However, most of the electronic medical record systems use International Classification of Diseases (ICD) codes for documenting diagnoses which are not sensitive enough for capturing specific drug-induced adverse reactions. Hence, direct identification of adverse reactions reliably from most of the currently available electronic medical record data sources is not feasible. In this study, we indirectly identified chemotherapy-induced adverse reactions from electronic medical records by conducting a causality analysis using readmission diagnoses data and drugs administered in the previously planned hospital admission and current medications.
As reported earlier [11], we also found an increased risk for readmission among patients who had drug-drug interactions. As established chemotherapy regimens were used, a majority of the patients had no drug-drug interactions while a relatively small proportion (17%) had not more than one interaction. Altogether, 29 chemotherapy regimens were used in the study population. However, only six regimens included drugs with clinically significant interactions. In all these interacting regimens, cisplatin interacted with either paclitaxel or docetaxel. It is established that platinum derivatives such as cisplatin enhance the myelosuppressive effect of taxane derivatives such as docetaxel and paclitaxel.
Our initial motivation for this investigation was to identify those factors that increase the risk for hospital readmission between two adjacent chemotherapy cycles. Upon completion of our analyses, we recognized that certain factors such as BMI, cisplatin dose, concomitant radiotherapy, drug-drug interactions, number of chemotherapy cycle, concomitant drugs, and cancer type can probably have a substantial impact on intercycle hospital admissions due to adverse reactions. However, age and gender did not have any influence on hospital readmission in our cohort. For clinical decision making the purpose, a more suitable approach for readers would be to consider consulting more similar studies preferably those using prospective methods to complement this data.
5. IMPLICATIONS FOR PRACTICE
Adverse reactions and unplanned hospital admissions among cancer patients have serious repercussions. Previous reports impart a significant number of patients either delaying or refusing further chemotherapy because of adverse reactions and related hospital readmissions [32]. One study has reported a significant reduction in survival rate among patients who delayed chemotherapy administration primarily due to adverse reactions [6]. Despite preventive steps including premedication, electrolyte preloading, clinical monitoring, and patient counselling, a significant number of patients receiving cisplatin-based chemotherapy regimen end up being readmitted due to adverse reactions. Identifying high-risk patient set would potentially aid the multidisciplinary healthcare team to plan, create or modify protocols and focus on further reducing the incidence of unplanned hospital admissions due to adverse reactions. Some of the associated factors found in this study such as cisplatin dose and drug-drug interaction are modifiable factors which imply a possible reduction of unplanned admission rate if proper attention is given to the high-risk patient group. Other associated factors such as BMI, concomitant radiotherapy and cancer type though are not modifiable, unplanned admissions could be potentially reduced by taking alternate measures (e.g. dose reduction). This could be of interest to hospital administrators as well considering the huge cost incurred by health systems or patients towards adverse reactions-related hospital admissions.
6. LIMITATIONS
This study has some limitations. This is a single center study in a tertiary care teaching hospital. The data sources used in this study are not comprehensive in data capturing. For example, information pertaining to any readmissions if occurred outside the study center were not included in the data source due to the lack of a national patient identifier system. However, to compensate for this limitation, we excluded patients with incomplete study data. This factor may have an effect on the final results. In addition, we did not analyze cancer stage, tumor size, or lymph node involvement, and radical intent of treatment of patients which are all associated with higher rates of readmission. Nevertheless, the association of these factors with increased risk of adverse reactions is not known.
CONCLUSION
Intercycle unplanned hospital admission due to adverse reactions was very common among patients receiving cisplatin-based chemotherapy regimen. The study findings highlight risk factors for readmission that may help identify high-risk patients who are more likely to be readmitted. In addition, the results suggest that therapy modifications based on associated factors coupled with other prophylactic clinical and non-clinical interventions may help reduce hospital readmission rate in these patients.
ACKNOWLEDGEMENTS
The authors are grateful towards Thenmozhi Mani and Andres Rodriguez for their expertise and assistance in particular with the statistical part of the study and for their help in writing the manuscript.
ETHICS APPROVAL AND CONSENT TO PARTICIPATE
This study was approved by the Institutional Review Board (IRB Min No: 10928 [Retro] dated 25.10.17).
HUMAN AND ANIMAL RIGHTS
Not applicable.
CONSENT FOR PUBLICATION
Not applicable.
AVAILABILITY OF DATA AND MATERIALS
The data that support the findings of this study are avalible from the corresponding author (H.B.R.) on reasonable request.
FUNDING
None.
CONFLICT OF INTERESt
The authors declare no conflict of interest, financial or otherwise.
REFERENCES
- 1.Dasari S., Tchounwou P.B. Cisplatin in cancer therapy: Molecular mechanisms of action. Eur. J. Pharmacol. 2014;740:364–378. doi: 10.1016/j.ejphar.2014.07.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Astolfi L., Ghiselli S., Guaran V., et al. Correlation of adverse effects of cisplatin administration in patients affected by solid tumours: A retrospective evaluation. Oncol. Rep. 2013;29(4):1285–1292. doi: 10.3892/or.2013.2279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Surendiran A., Balamurugan N., Gunaseelan K., Akhtar S., Reddy K., Adithan C. Adverse drug reaction profile of cisplatin-based chemotherapy regimen in a tertiary care hospital in India: An evaluative study. Indian J. Pharmacol. 2010;42(1):40–43. doi: 10.4103/0253-7613.62412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sumiyoshi M., Soda H., Sadanaga N., et al. Alert regarding cisplatin-induced severe adverse events in cancer patients with xeroderma pigmentosum. Intern. Med. 2017;56(8):979–982. doi: 10.2169/internalmedicine.56.7866. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Pujol J-L., Lafontaine T., Quantin X., et al. Neoadjuvant Etoposide, Ifosfamide, and Cisplatin followed by concomitant thoracic radiotherapy and continuous cisplatin infusion in stage IIIb non-small cell lung cancer. Chest. 1999;115(1):144–150. doi: 10.1378/chest.115.1.144. [DOI] [PubMed] [Google Scholar]
- 6.Otty Z., Skinner M.B., Dass J., et al. Efficacy and tolerability of weekly low-dose cisplatin concurrent with radiotherapy in head and neck cancer patients. Asia Pac. J. Clin. Oncol. •••;7(3):287–292. doi: 10.1111/j.1743-7563.2011.01405.x. [DOI] [PubMed] [Google Scholar]
- 7.Felix H.C., Seaberg B., Bursac Z., Thostenson J., Stewart M.K. Why do patients keep coming back? results of a readmitted patient survey. Soc. Work Health Care. 2015;54(1):1–15. doi: 10.1080/00981389.2014.966881. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bell J.F., Whitney R.L., Reed S.C., et al. Systematic review of hospital readmissions among patients with cancer in the United States. Oncol. Nurs. Forum. 2017;44(2):176–191. doi: 10.1011/17.ONF.176-191. [DOI] [PubMed] [Google Scholar]
- 9.Donzé J., Lipsitz S., Bates D.W., Schnipper J.L. Causes and patterns of readmissions in patients with common comorbidities: Retrospective cohort study. BMJ. 2013;•••:347. doi: 10.1136/bmj.f7171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ji H., Abushomar H., Chen X-K., Qian C., Gerson D. All-cause readmission to acute care for cancer patients. Healthc. Q. 2012;15(3):14–16. doi: 10.12927/hcq.2013.23044. [DOI] [PubMed] [Google Scholar]
- 11.El Morabet N., Uitvlugt E.B., van den Bemt B.J.F., van den Bemt P.M.L.A., Janssen M.J.A., Karapinar-Çarkit F. Prevalence and preventability of drug-related hospital readmissions: A systematic review. J. Am. Geriatr. Soc. 2018;66(3):602–608. doi: 10.1111/jgs.15244. [DOI] [PubMed] [Google Scholar]
- 12.Steventon A., Billings J. Preventing hospital readmissions: The importance of considering ‘impactibility,’ not just predicted risk. BMJ Qual Amp Saf. 2017;26(10):782. doi: 10.1136/bmjqs-2017-006629. [DOI] [PubMed] [Google Scholar]
- 13.2011 https://www.cms.gov/Medicare/ Quality-Initiatives-Patient-Assessment-Instruments/MMS/downloads/MMSHospital-WideAll-ConditionReadmissionRate.pdf
- 14. http://www.who.int/medicines/areas/quality_safety/ safety_efficacy/WHOcausality_assessment.pdf
- 15.Brana I., Siu L.L. Locally advanced head and neck squamous cell cancer: Treatment choice based on risk factors and optimizing drug prescription. Ann. Oncol. 2012;23(10):x178–x185. doi: 10.1093/annonc/mds322. [DOI] [PubMed] [Google Scholar]
- 16.Brown E., Burgess D., Li C., Canter R., Bold R. Hospital readmissions: Necessary evil or preventable target for quality improvement. Ann. Surg. 2014;260(4):583–591. doi: 10.1097/SLA.0000000000000923. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Shapiro J.S., Humeniuk M.S., Siddiqui M.A., Bonthu N., Schroeder D.R., Kashiwagi D.T. Risk factors for readmission in patients with cancer comanaged by hospitalists. Am. J. Med. Qual. 2017;32(5):526–531. doi: 10.1177/1062860616665904. [DOI] [PubMed] [Google Scholar]
- 18.Solomon R., Egorova N., Franco R., Bickell N.A. Thirty-day readmissions in metastatic cancer patients: Room for improvement? J. Clin. Oncol. 2017;35(15):6541. doi: 10.1200/JOP.18.00500. [DOI] [PubMed] [Google Scholar]
- 19.Epstein A.S., Crosbie C., Martin S.C., et al. 30-day-or-sooner readmissions of gastrointestinal medical oncology patients following cancer center inpatient service discharge: Characteristics and preventability. Hosp. Pract. 2014;42(5):34–44. doi: 10.3810/hp.2014.12.1157. [DOI] [PubMed] [Google Scholar]
- 20.Ghiam M.K., Langerman A., Sargi Z., Rohde S. Head and neck cancer patients: Rates, reasons, and risk factors for 30-day unplanned readmission. Otolaryngol. Head Neck Surg. 2018;159(1):149–157. doi: 10.1177/0194599818776633. [DOI] [PubMed] [Google Scholar]
- 21.Berry J.G., Gay J.C., Joynt M.K., et al. Age trends in 30 day hospital readmissions: US national retrospective analysis. BMJ. 2018;360:•••. doi: 10.1136/bmj.k497. http://www.bmj.com/content/360/bmj.k497 [abstract]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.González J.R., Fernandez E., Moreno V., et al. Sex differences in hospital readmission among colorectal cancer patients. J. Epidemiol. Community Health. 2005;59(6):506. doi: 10.1136/jech.2004.028902. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Chang H-T., Chen C-K., Lin M-H., Chou P., Chen T-J., Hwang S-J. Readmissions in cancer patients after receiving inpatient palliative care in Taiwan: A 9-Year Nationwide population-based cohort study. Medicine (Baltimore) 2016;95(8):e2782. doi: 10.1097/MD.0000000000002782. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Sebastian M.R., Hergenroeder A. Does rate of weight gain during hospitalization predict readmission in adolescents with eating disorders? J. Adolesc. Health. 2014;54(2):S78–S79. doi: 10.1515/ijamh-2018-0228. [DOI] [PubMed] [Google Scholar]
- 25.Reinke C.E., Kelz R.R., Zubizarreta J.R., et al. Obesity and readmission in elderly surgical patients. Surgery. 2012;152(3):355–362. doi: 10.1016/j.surg.2012.06.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zai A.H., Ronquillo J.G., Nieves R., Chueh H.C., Kvedar J.C., Jethwani K. Assessing hospital readmission risk factors in heart failure patients enrolled in a telemonitoring program. Int. J. Telemed. Appl. 2013;2013:305819. doi: 10.1155/2013/305819. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Andersen K.K., Olsen T.S. The obesity paradox in stroke: Lower mortality and lower risk of readmission for recurrent stroke in obese stroke patients. Int. J. Stroke. 2013;10(1):99–104. doi: 10.1111/ijs.12016. [DOI] [PubMed] [Google Scholar]
- 28.Manzano J-G.M., Gadiraju S., Hiremath A., Lin H.Y., Farroni J., Halm J. Unplanned 30-Day readmissions in a general internal medicine hospitalist service at a comprehensive cancer center. J. Oncol. Pract. 2015;11(5):410–415. doi: 10.1200/JOP.2014.003087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Bernier J., Domenge C., Ozsahin M., et al. Postoperative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N. Engl. J. Med. 2004;350(19):1945–1952. doi: 10.1056/NEJMoa032641. [DOI] [PubMed] [Google Scholar]
- 30.Wolfe D., Yazdi F., Kanji S., et al. Incidence, causes, and consequences of preventable adverse drug reactions occurring in inpatients: A systematic review of systematic reviews. PLoS One. 2018;13(10):e0205426. doi: 10.1371/journal.pone.0205426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Min S., Zheng Q., Zhang B., et al. A meta-analysis of efficacy and adverse effects of lobaplatin and cisplatin in the treatment of malignant pleural effusion. Cancer. 2019;22(2):90–98. doi: 10.3779/j.issn.1009-3419.2019.02.03. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Suh W.N., Kong K.A., Han Y., et al. Risk factors associated with treatment refusal in lung cancer. Thorac. Cancer. 2017;8(5):443–450. doi: 10.1111/1759-7714.12461. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are avalible from the corresponding author (H.B.R.) on reasonable request.