Association of serum calcium concentrations with renal impairment in Chinese patients with newly diagnosed multiple myeloma: A Retrospective Study

Renal impairment(RI) is a common complication of multiple myeloma. Few studies have been conducted to determine the association between the serum calcium concentration and the occurrence of RI in MM patients. In this retrospective study, we included 568 patients with newly diagnosed MM who participated in an ongoing retrospective cohort study. The serum calcium concentrations and the presence of RI at baseline in MM patients were evaluated. Patient data were collected at baseline and multiple regression analyses were used to estimate independent relationships. We further used a two piecewise linear regression model to identify non-linear relationships. independent was fully analyses The test for interactions was not statistically significant. serum was independent associated with the presence of RI in patients with MM. There was a nonlinear relationship between the serum calcium concentration and the presence of RI. The serum calcium concentration was positively related with the presence of RI when the albumin-adjusted serum calcium concentration was >2.3mmol/L. Our study suggested that we should take measures to reduce the blood calcium concentration earlier rather than waiting for hypercalcemia to occur.


Introduction
Multiple myeloma (MM) is a clonal B-cell malignancy of the bone marrow that is associated with a variety of clinical manifestations, including hypercalcemia, renal impairment (RI), anemia, and bone disease. MM is the second most common hematologic malignancy and accounts for 1% of all malignancies (1).
RI is a common complication of MM. Depending on the definition of RI (defined as a serum creatinine level > 2 mg/dL), this complication is reported in 15-40% of patients with MM. At the time of diagnosis, 30-40% of patients with MM have a serum creatinine level above the upper limit of normal and the majority have a serum creatinine < 4 mg/dL (2).
Despite promising evidence on hypercalcemia (defined as a serum calcium concentration > 11.5 mg/dL or 2.85 mmol/L) as an important biomarker of MM RI (3)(4),these studies lack details on the effect of the serum calcium concentration on the serum creatinine level and the presence of RI. No interaction and quantitative analyses were performed. Also the study lacks the exploration and presentation of a linear relationship between the serum calcium concentration with MM-related RI occurrence. Therefore, we propose a clinical hypothesis that there is a correlation between serum calcium concentration and MM-related RI, but an increase in serum calcium concentration will lead to MM-related RI before reaching hypercalcemia. So this study aimed to assess the a linear and non-linear relationship between the serum calcium concentration and the presence of RI within the Chinese population for early prevention of kidney damage. The influence of other risk factors, such as anemia, immunoglobulin levels, and light chain protein was also evaluated by using interaction and stratified analyses.

Subjects and methods
In this single center retrospective study was on the basis of a subset of an ongoing retrospective cohort study, we collected clinical and hematological data for newly

Participants
The inclusion criteria were as follows: 1) newly diagnosed symptomatic MM patients without a history of prior chemotherapy for MM; 2) no history of other solid tumors; and 3) no radiotherapy.
The exclusion criteria were as follows: 1) no integrated data for effects, such as serum albumin and serum total calcium; and 2) a history of kidney disease, severe infection, liver disease, or autoimmune disease.
The specific details of enrollment and exclusion are shown in the following flow chart (Fig. 1).

Evaluated indicators
The following indicators were evaluated: 1) demographic characteristics (age, sex, and underlying disease [hypertension and diabetes]); 2) baseline data, including routine blood and serum biochemical testing, immunoglobulin concentrations, routine urinalysis, and other data pre-chemotherapy; 3) hypercalcemia, defined as a corrected serum calcium greater than an elevated serum calcium (> 2.85 mmol/L or 11.5 mg/dL); 4) the presence of RI, defined as a serum creatinine level (2 mg/dL or177 umol/L) at the time of diagnosis [5,6]; 5) ISS classification was used for staging all patients [7,8].

Statistical analysis
Continuous data are expressed as the mean ± standard deviation and median (interquartile range). The categorical variables are presented as a number or percentage. The difference between two groups was assessed using a Student's ttest, chi-square test, or Mann-Whitney U test, as appropriate. The trends of variables were detected by the linear-by-linear chi-square test.
We then used a univariate linear regression model analyses to estimate the independent relationship between the albumin-adjusted serum calcium concentration and the risk of MM-related RI with an adjustment for potential confounders. Both non-adjusted and multivariate adjusted models are listed.
According to the recommendation of the STROBE statement, we simultaneously showed the results of unadjusted, minimally adjusted, and fully adjusted analyses.
Whether or not the covariances were adjusted was determined by the following principle: when added to this model, the matched odds ratio changed by at least 10%. We further used a two piecewise linear regression model to identify the nonlinear relationship. If a nonlinear correlation existed, a two piecewise linear regression model was used to calculate the threshold effect of the calcium concentration on MM-related RI in terms of the smoothing plot. When the threshold level (i.e., turning point) was apparent on the smoothed curve, the inflection point was automatically calculated by the recursive method and the maximum model likelihood was used [12]. Subgroup analyses were performed using stratified linear regression models. The modification and interaction of the subgroup were inspected by the likelihood ratio test. Statistical analysis was performed using Empower States (www.empowerstats.com; X & Y Solution, Inc., Boston, MA, USA) and R software (http://www.R-project.org). A P value < 0.05 was considered significant.

Baseline characteristics
Of the 602 participants, 33 were excluded from this study, thus leaving 568 subjects for data analysis (Fig. 1). Based on a serum creatinine level > 2 mg/dL or 177 umol/L, we observed that 155 of 568 patients(27.2%)with newly diagnosed MM presented with the presence of RI, this percentage was similar to the incidence of renal failure in patients.
The characteristics of the study population are listed in Table 1, and Fig. 2 shows the distribution of the mean albumin-adjusted serum calcium concentration based on the presence of RI. Using the definition of IMWG(International Myeloma Working Group), [please spell out with 1st use] 11.4% of the patients with symptomatic MM had hypercalcemia (i.e., a corrected serum calcium ≥ 11.5 mg/dL) at the time of diagnosis.     As shown in Table 5, the test for interactions was not statistically significant for age, sex, Hemoglobin, LDH, and serum albumin (P for interaction > 0.05; Table 5).

Discussion
In this study we demonstrated that the serum calcium concentration was associated with the serum creatinine level and MM-related RI occurrence in Chinese patients with newly diagnosed MM in basic or fully adjusted analyses. Hypercalcemia was associated with a high risk of RI occurrence in MM patients. We also showed a   (17,18).
Hypercalcemia(defined as a serum calcium concentration > 11.5 mg/dL or 2.85 mmol/L)) is as the second most common cause of renal failure in patients with MM, except free light chains (19,20). Some researchers are of the opinion that hypercalcemia and/or Bence-Jones proteinuria explain renal failure in 97% of patients (3). Hypercalcemia interferes with renal function and impairs the renal concentrating ability, causes vasoconstriction of the renal vasculature and enhances diuresis, which may result in hypovolemia and pre-renal azotemia. Concentrated urine and reduced urine flow enhance cast formation, thus leading to further renal damage (19,20).
Hypercalcemia is a defining characteristic of symptomatic MM, and observed approximately in 15% of newly diagnosed patients (4,21). Our cohort came from China and showed a low incidence of hypercalcemia (11.4%).
Hypercalcemia is an important cause of renal failure, but whether or not hypercalcemia is an independent risk factor for loss of renal function is controversial. Zagouri et al. (7) reported that hypercalcemia is associated with a lower Estimating Glomerular Filtration Rate (eGFR) as an independent risk factor, but logistic regression analysis was then performed at the University of Athens School of Medicine, which showed that creatinine is independently associated with ISS stage and Bence-Jones proteinuria (21).Our study showed hypercalcemia was independently associated with a high risk of RI occurrence in MM patients.
Colleagues should also pay attention to a problem which is very important to correct the serum calcium level with albumin. Most of the protein-bound calcium is bound to albumin, while the remainder is complexed to globulins. Disorders that lower serum albumin will lower total serum calcium, but have a lesser effect on the ionized calcium concentration (9)(10)(11). The guidelines for MM also recommend the use of the albumin-adjusted serum calcium concentration as a research indicator (6).
The main problem of the existing researchs is that these studies were only concerned about the effects of hypercalcemia on MM-related RI occurrence (4,7,21).
At the same time, the calcium level in these studies were not corrected by serum albumin, and till this moment few similar studies were conducted to investigate the association between the serum calcium concentration and MM-related RI occurrence in patients with MM. Therefore, we propose a clinical hypothesis that there is a correlation between serum calcium concentration and MM-related RI, but an increase in serum calcium concentration will lead to MM-related RI before reaching hypercalcemia.
The occurrence of renal damage is closely related to the serious prognosis of MM.
Our study confirmed that the serum calcium concentration was associated with the serum creatinine level and the occurrence of MM-related RI ,and hypercalcemia is an independent factor in renal damage in Chinese patients with newly diagnosed MM. More importantly, we found that there is a nonlinear relationship between the albumin-adjusted serum calcium concentration and the occurrence of RI. There was a positive correlation between the serum calcium concentration and the occurrence of RI when the serum calcium concentration was > 2.3 mmol/L (p < 0.05); however, these associations were no longer statistically significant when the serum calcium concentrations was < 2.3 mmol/L in fully adjusted analyses (p > 0.05). Our study suggested that we should take measures to lower the blood calcium level to prevent kidney damage when the albumin-adjusted serum calcium concentration exceeds 2.3 mmol/L rather than after the occurrence of hypercalcemia.
The limitations of our study included the observational design, retrospective ascertainment of the serum calcium concentration, the serum creatinine level at a single time point late during the course of newly diagnosed MM, and the potential misclassification of study measurements.

Conclusions
In conclusion, our study showed that the serum calcium concentration was independent associated with the occurrence of RI and serum creatinine level in Chinese patients with newly diagnosed MM. There was a nonlinear relationship between the serum calcium concentration and the occurrence of RI. The serum calcium concentration was positively related with the occurrence of RI the when albumin-adjusted serum calcium concentration was > 2.3 mmol/L. Our study suggested that we should take measures to reduce blood calcium earlier, instead of waiting for hypercalcemia to occur.  The relationship between serum calcium level and the occurrence of Myeloma-Related RI. A nonlin