6 курс / Нефрология / Острое_повреждение_почек_после_паратиреоидэктомии_по_поводу_первичного

252

In the first phase, the risk factors were analyzed in a group of patients with normal renal function.

A logistic regression model was built, which was statistically significant: χ2= 28.86, df=8, p<0.001, RN2=0.38. Regression coefficients exponents and their statistical significance are shown in Table 4.19.

Table 4.19. Full Model (logistic regression) of the preoperative factors and the AKI risk in patients with preserved renal function (eGFR≥60 mL/min/1.73 m2).

The model presented above has acceptable prognostic value: the area under the ROC curve of the values predicted by the model was 0.807 [95%CI 0.708; 0.905], p<0.001 (Figure 4.2).

This model was also simplified. As a result, a logistic regression model was obtained, which was statistically significant: χ2= 25.39, df= 5, p<0.001, RN2= 0.341. Regression coefficients exponents and their statistical significance are shown in Table 4.20.

Table 4.20. Reduced model (logistic regression) of the preoperative factors and the

AKI risk patients with preserved renal function (eGFR≥60 mL/min/1.73 m2).

254

The optimal cut-off point of the AKI predicted probability in the full model was >0.602, which corresponded to the Youden’s index of 0.536, in the reduced model > 0.58 and 0.525, respectively.

This cut-off point for the reduced model provides AUC-ROC 0.762 [95%CI 0.652; 0.872], Se 0.583 [95%CI 0.422; 0.729], Sp 0.941 [95%CI 0.841; 0.984], p<0.0001.

One of the study goals was to stratify patients according to the AKI risk. From the critical evaluation of the reduced model presented in Table 4.20, it becomes apparent preoperative PTH and total serum calcium levels are the modifiable risk factors. Quantitative risk factors are eGFR, PTH and total calcium levels, which are measured in different units, thus do not allow direct comparison of their changes effect on the AKI risk. However, it is important to assess the change in the AKI risk per unit of quantitative predictors variability. To overcome this limitation, the procedure for predictors standardization was applied:

Standardized predictor value = − ,

where is the observed predictor value, is the mean predictor value, SD is the standard deviation.

The following estimates of standardized predictors were obtained - Table 4.21. Thus, the effect of different predictors expressed in the same units on the AKI

probability was estimated (i.e. the change in the AKI probability per unit of predictor variability).

Table 4.21. Reduced model (logistic regression) of the standardized predictors and the AKI risk in patients with preserved renal function (eGFR≥60 mL/min/1.73 m2).

258

Clinical observation 2.

Patient M., female, 50 years old, was hospitalized for routine PTx for PHPT. Baseline eGFR was 89.2 mL/min/1.73 m2 (creatinine level 68.8 mmol/L). Anamnesis: constant intake of ACEi for hypertension stage 2, iron deficiency anemia, pronounced PHPT bone manifestations (low-energy fractures, "brown" tumors, osteopenia according to densitometry). At the time of admission, severe hypercalcemia was noted: total serum calcium level 3.78 mmol/L, ionized calcium level – 2.16 mmol/L. Preoperative PTH level 433 pmol/L, alkaline phosphatase level 518 U/L. Before the surgery, the patient was in the ICU for 3 days, where she was being prepared for the surgery (intravenous hydration, correction of electrolyte disorders were performed). A selective PTx was performed, the right lower PTG was removed of 30x25x20 mm size and of 6.8 g weight. The surgery main stage duration of 30 minutes, the ventilation duration of 85 minutes, intraoperative hypotension was not recorded. In the postoperative period, there was a PTH level decrease (0.8 pmol/L) and a slow ionized blood calcium level decrease (1.86 -1.45-1.4 mmol/L), and the AKI stage 2 with oliguria according to KDIGO-2012 criteria occurred within 24-48 hours after the PTx (maximum creatinine level 128 mmol/L), which required patient management and treatment in the ICU for 1 day. The patient was discharged on the 11th day with improvement, however with no complete renal function recovery by the time of discharge, an outpatient consultation with a physician/nephrologist was recommended.

Clinical observation 3.

Patient A., female, 62 years old, was admitted for routine PTx for PHPT. Baseline eGFR was 96 mL/min/1.73 m2 (creatinine level 56 mmol/L). Anamnesis: AH stage 3, grade 2, cardiovascular complications risk grade 3, hypotensive therapy intake (ACEi/ARB). Concomitant diseases: CHF II functional class (NYHA), COPD, urolithiasis, osteopenia according to densitometry. The comorbidity index of 10 points per the CIRS. Laboratory: preoperative PTH level - 54.7 pmol/L, total blood

260

RESUME

Primary hyperparathyroidism is a widespread socially important endocrine disease, caused in most cases by benign tumor (adenoma) of the parathyroid gland. PHPT is characterized by increased parathyroid hormone levels in combination with normal or elevated blood calcium levels. The disease manifestations can be poor and non-specific for a long time, with this the absence of timely treatment leads to various organs and systems disorders, which also confuse the diagnosis, and often result in complications development and patients’ disability.

Primary hyperparathyroidism is the third most common endocrine disease affecting mainly postmenopausal women. The prevalence of PHPT according to different authors varies widely [54, 190, 207] that is due to the medical care is not equally available and not of the same quality in countries at different stages of economic progress. In developed countries, with routine biochemical examination available for the population, PHPT is generally identified more often, thus "mild" forms dominate with no prominent clinics. Manifesting forms with target organs (kidneys, bone tissue) damage and pronounced hypercalcemia remain dominant in less economically developed countries including Russia [2].

The main and most effective treatment for PHPT is surgery. Parathyroidectomy leads to osteoporosis and neurocognitive disorders recovery, and reduces the risk of fractures and nephrolithiasis [118]. Surgical treatment reaches 98% effectiveness with a comparatively low incidence of postoperative complications if performed by an experienced surgeon [100]. Since PTx is a relatively low-invasive operation, it does not imply significant postoperative complications; some countries even allow this surgery to be performed on an outpatient basis [20]. Well studied complications of PTx include early surgical complications (bleeding, postoperative wound infection, recurrent laryngeal nerve paresis) and late ones (relapse, persistence). The main "therapeutic" complication of PHPT surgical treatment and associated with the disease is postoperative hypocalcemia, which is often caused by hungry bone syndrome due to previous hyperparathyroid osteodystrophy. Little is known about the other therapy complications