Surgical complications of peritoneal dialysis in children with acute kidney failure

Full Text

INTRODUCTION

Acute renal failure (ARF) is a polyetiological syndrome characterized by an acute impairment of a renal function, leading to a homeostasis disorder. ARF typically develops over a period of several hours to several days in response to various injuries and is evident through hyperhydration, azotemia, and violation of the acid–base state and electrolyte balance. A special feature of the course of ARF is its cyclical nature with the possibility of a complete recovery of renal functions [3].

Although ARF is diagnosed in about 3%–5% of all patients in intensive care units and critical care medicine, the frequency of ARF among neonates requiring dialysis is only 1 in 5,000 newborns and is generally caused due to the malformations of the urinary system, congenital heart defects, and prerenal (dehydration, bleeding, sepsis, and anorexia) and renal factors. The incidence of ARF among children aged six months to five years, however, goes up to 4–5 cases per 10,000 children, the main cause being hemolytic–uremic syndrome (HUS). Interestingly, about 55%–70% of the ARF cases occur due to HUS; however, with timely and proper treatment, the kidney function is fully restored in most cases [4, 5, 9].

Further, a peritoneal dialysis (PD) is the preferred method of treatment for children with ARF. Despite the widely used measures to improve the results of PD, complications are extremely common.

Although since the 1990s, PD has been commonly used in the Dialysis Department as a method of a renal replacement therapy for an ARF of children, since the 2010s, as the methodology became more developed, and algorithms for surgical tactics have been developed, it became the main method. Nowadays, the implantation of a peritoneal catheter for children who need an urgent dialysis is performed by an open method through minilaparotomic access. A prerequisite is the resection of the omentum and the creation of a subcutaneous tunnel. In our hospital (i.e., the Children Municipal Multidisciplinary Clinical Specialized Center of High Medical Technologies), standard catheters of the Tenckhoff 210 and 516 models with two cuffs are used.

Further, according to the data provided for 2011 by the North American Pediatric Renal Trials and Collaborative Studies, catheter revision was required in 19% of all implants. The reasons for the revision were catheter dysfunction (40%), dialysis peritonitis (16%), infection of the catheter exit point (14%), dialysate leakage (4%), and others (26%) [12].

Noninfectious complications of PD include outflow disorders (5%–24%), dialysate leaks (7%–10%), anterior abdominal wall hernias (8%), intra-abdominal bleeding (7%), pain when filling and draining the solution (4%), catheter migration (3%), damage to the catheter by the patient or a healthcare provider (3%), an increased abdominal size (1.5%), erosion above the cuff and its extrusion, and granulation of the catheter exit site [2, 6, 8, 10].

Rare complications, on the other hand, include sclerosing encapsulating peritonitis, hydrothorax, hydropericardium, pancreatitis, ischemic colitis, necrotic enterocolitis, and pneumoperitoneum [7, 11]. Infectious problems remain the most common cause of children’s morbidity at chronic PD, and the frequency of peritonitis for children exceeds that for adults [12].

The treatment of surgical complications depends on the experience and qualifications of the attending physician. Early diagnosis and intervention plays a significant role in reducing the frequency of catheter abnormalities and patient mortality with PD.

The aim of this work was to study the causes and structure of surgical complications of PD in order to develop possible methods of the prevention and improvement of treatment results for children with ARF getting PD.

MATERIALS AND METHODS

This retrospective study analyzed the medical histories of 57 patients with an ARF who were treated in the Dialysis Department of the Children Municipal Multidisciplinary Clinical Specialized Center for High Medical Technologies in St. Petersburg and who received PD as a method of renal replacement therapy during 2008–2018. The Dialysis Department has been functioning since 1977, and it is one of the leading specialized departments in the North-Western region of the Russian Federation providing care for children with both acute and chronic kidney failure.

The age of the children included in this study ranged from seven months to seven years (the average age was 2.7 years ± 2.5 months). Although the inclusion criterion of the study was the diagnosis of an isolated ARF (acute kidney failure in the structure of multiple organ failure was an exception), newborns who received PD were not included as this method has been used for this age group since 2017, and the accumulated experience does not yet allow us to draw conclusions about the frequency and qualitative composition of PD complications. Of the 57 children included in the study, 28 (49%) were boys and 29 (51%) girls.

In the current study (i.e., in the age group of seven months to seven years), although HUS was predictably the leading cause of ARF for children in 49 cases (86%), thrombotic thrombocytopenic purpura (Moskowitz syndrome) was found in 2 cases (3.5%), interstitial nephritis in 2 (3.5%), Reye’s syndrome in 1 (1.75%), acute intravascular post-transfusion hemolysis in 1 case (1.75%), dehydration due to overheating (high ambient temperature) with organic damage of the central nervous system and psychomotor development delay in 1 case (1.75%), and acute glomerulonephritis in 1 case (1.75%).

In this retrospective study, the results of patients with surgical complications of PD were processed on a personal computer using the Statistica 13.3 application software package. This software performs all calculations using standard mathematical statistics formulas, using only existing, measured data (all omissions were excluded from the calculations and not taken into account when forming conclusions). Moreover, Statistica 13.3 allows one to perform all classic types of analysis using an extremely wide set of specific algorithms and methods.

The array of initial data for this work was prepared to form groups and subgroups in accordance with various criteria and perform calculations at the desired level of detail. While the normality of the sample distribution was assessed using the Kolmogorov–Smirnov test, the reliability of the obtained data was determined using the Fischer’s exact, Mann–Whitney, and chi-squared tests.

In addition, p < 0.05 was considered as the level of statistical significance.

RESULTS AND DISCUSSION

Interestingly, among the 57 children who received PD as renal replacement therapy for ARF, 37 (65%) had no dialysis complications.

The average time for an implantation of a peritoneal catheter was 45 min (±3 min), the minimum time was 20 min, and the maximum time was 105 min. Besides, the resection of the omentum was made in 37 cases (65%) and is explained by the fact that until 2012, omentectomy in our hospital was optional.

Indications for the beginning of PD and implantation of a peritoneal catheter occurred on average on the seventh day (±0.7 s) from the onset of the main disease (most often gastroenteritis). We considered the period from the onset of the disease to the development of ARF requiring dialysis therapy as a possible factor for predicting the duration of dialysis. In this case, no statistically reliable relationship was observed between the period of the onset of the disease and duration of PD (p > 0.05).

Since the indications for the beginning of a renal replacement therapy for acute kidney failure are put on an urgent basis, the period recommended in clinical guidelines from the implantation of a peritoneal catheter to the beginning of dialysis (14 days) is not maintained for obvious reasons. In our study, 93% of the patients started receiving dialysis a few hours after the implantation of the peritoneal catheter, while 7% started receiving it on the first day.

The duration of PD averaged 13.3 days (±1.1 s), the minimum period was 2 days, and the maximum was 43. There was no link between the duration of dialysis and the number of complications as well as the number of cases requiring a surgery (p > 0.05).

PD complications for children with ARF were 35%, and 22.8% of PD patients required a surgical treatment. Table 1 presents a complete list of complications.

 

Table 1 / Таблица 1 The structure of surgical complications of peritoneal dialysis in children with acute renal failure Структура хирургических осложнений перитонеального диализа у детей с острой почечной недостаточностью
Type of complication / Вид осложнения	Patients / Пациенты	Percentage, % Доля, %	The number of operated / Количество оперированных
Catheter dysfunction / Дисфункция катетера	10	17.5	8
Pericatheter leakage / Перикатетерные утечки	8	14.0	4
Pain syndrome / Болевой синдром	2	3.5	2
Catheter bleeding / Кровотечение из места выхода катетера	1	1.7	0
Patent processus vaginalis / Патология влагалищного отростка	1	1.7	1
Intestinal Invagination / Тонко-тонкокишечная инвагинация	2	3.5	2
Peritonitis / Перитони	2	3.5	0

 

Of note, one patient may have several different complications (e.g., dysfunction in combination with pain, pericatheter leaks and peritonitis, or pericatheter leaks after a surgical revision for catheter dysfunction); another patient may have undergone more than one surgical intervention.

Nevertheless, at the same time, out of all 20 children with PD complications, 13 (65%) required a surgery, which accounted for 22.8% of the total number of patients on dialysis. Let us focus on each complication separately.

Pericatheter leaks of dialysate–as one of the most expected complications of PD for children with ARF due to a short introductory period (and more often almost without it)–occurred in eight patients (14% of the total group). The time for diagnosing leaks is shown in Fig. 1.

 

The time of pericatheter leakage manifestation in children receiving peritoneal dialysis for acute renal failure

Время появления перикатетерных протечек у детей, получающих перитонеальный диализ по поводу острой почечной недостаточности

 

Although four of the eight patients (50%) underwent a surgical treatment for dialysate leaks, others were chosen for conservative tactics. The latter were stopped on the second, fourth, and seventh days, and dialysis was not resumed for all four of them (neither peritoneal nor hemodialysis). In addition, in one case, pericatheter leaks were not an indication for immediate discontinuation of dialysis, but the dialysis was continued despite the risk of developing peritonitis. Analyzing the surgical activity, two of the four patients underwent the revision of the catheter implantation site and the suturing of the peritoneal defect, where one was a reimplantation of the catheter, and the other was a removal of the catheter with the termination of PD (since this operation was the third in four days, and during the intervention, it was diagnosed with a thin-intestinal invagination). Thus, pericatheter leaks are a frequent complication that affects the duration of PD [five (62.5%) patients stopped receiving dialysis], and an acute need for dialysis requires active tactics on the part of the surgeon.

Catheter migration and obstruction constitute the concept of catheter dysfunction due to mechanical causes. In our study, catheter dysfunction was found in 17.5% of 10 patients, and 2 patients showed a slowdown in draining, which neither required a surgery nor affected the timing of dialysis. However, a surgery was required in eight other cases.

Revision of the catheter due to its dysfunction was made on average on the third day after the implantation of the catheter and the beginning of dialysis (at least on the first, at most on the eleventh). Table 2 presents the causes of a dysfunction.

 

Table 2 / Таблица 2 Causes of peritoneal catheter dysfunction in children with acute renal failure Причины дисфункции перитонеального катетера у детей с острой почечной недостаточностью
Cause of dysfunction / Причина дисфункции	Patients / Пациенты	Percentage, % / Доля, %
Colon hydatide obturation / Обтурация жировыми подвесками толстой кишки	4	40
Omental obstruction / Обтурация сальником	1	10
Fimbria of the fallopian tubes obstruction / Обтурация фимбриями маточных труб	1	10
Catheter transposition / Миграция катетера	2	20
Slow drain / Замедление слива	2	20
Total / Всего	10	100

 

Besides, cases of the obturation of the catheter by the omentum appear to be a frequent phenomenon, and in our group of patients, as mentioned earlier, omentectomy for prophylactic purposes was made for almost all patients. However, of those who did not have it, only one had catheter dysfunction due to omental wrapping. Fimbriae of the fallopian tubes, as the cause of catheter obturation, have received a little attention in the literature, which in our opinion, should be studied more since this phenomenon is common and fairly difficult to prevent. Another difficult-to-prevent complication is the obturation of the catheter with the fat suspensions of the colon, more often sigmoid. As a rule, they enter the lumen of the catheter through small side perforations, necrotize, and completely obturate the lumen. Of the four children who underwent catheter revision for obturation with fat suspensions, one had to undergo three surgeries for four days. After the traditional implantation of the catheter with an omentectomy, the first revision of the catheter was made on the second day, the catheter was obturated with colon suspensions and reimplanted (the cause was a large peritoneal defect). The next day (the third day from the start of dialysis), due to the lack of outflow, indications were given for a second revision, another obturation of the catheter with suspensions was detected, as well as a thin-intestinal invagination, which was straightened out without technical difficulties. The catheter was reimplanted (also due to a large peritoneal defect). PD was resumed, but there were pericatheter leaks of dialysate, the intensity of which increased by the sixth day when the third revision was performed, wherein another thin-bowel invagination was detected, and the catheter was removed. Later, the child’s diuresis was restored, and he was discharged home. The last case clearly shows how difficult it is to prevent catheter obturation with fat suspensions and how unpredictable the frequency of catheter revisions can be. It is also worth noting that access during catheter revision should be done separately from access during catheter implantation (e.g., over the womb) in order to maintain tightness at the point where the catheter enters the abdominal cavity. Otherwise, each revision may end with a catheter reimplantation.

The literature discusses the possibility of a laparoscopic revision of the catheter in its dysfunction. Our experience shows that during laparoscopy, it is difficult to achieve the tightness of the peritoneum for a child on the first day, which is necessary for the resumption of dialysis. In one patient who underwent a laparoscopic revision (obturation of the catheter with fat suspensions) and sealing of the peritoneal operating wounds, there was a leak of dialysate from postoperative wounds on the second day after the resumption of dialysis, which required the suspension of dialysis. Thus, despite the minimally invasive and cosmetic access during laparoscopy, minilaparotomy for the revision of catheter dysfunction in children with ARF is currently preferred.

Catheter migration is an infrequent complication of children. In our study, it was observed in two patients, both of whom underwent reposition of the catheter into the pelvic cavity surgically. Conservative measures (cleansing enemas to activate intestinal motility) were ineffective. In one case, the reason for the migration of the catheter was the wrong choice of the length of the catheter by the surgeon on duty: a very long intra-abdominal part descended into the pelvis and made a bend there, and the end ended up in the left hypochondrium. We performed the revision and shortening of the catheter, after which the latter “laid” in the pelvic cavity without difficulties. The second child had the migration that was caused by an incorrectly oriented subcutaneous tunnel, which in the conditions of short dialysis periods for acute kidney failure, did not require its plasticization but only the reposition of the catheter into the pelvic cavity.

Three patients had such a rare complication for PD in ARF as peritonitis. In one case, the cause was pericatheter leaks, in the second, it was the failure to detect the predisposing factor, and in the third case, peritonitis was diagnosed on the 85th day after the implantation of the catheter. By this time, the child had not received dialysis for more than a month, but the catheter was not removed due to chronic renal failure and borderline indicators of azotemia. The catheter was removed on the 85th day due to the development of peritonitis. The child continued to be monitored by nephrologists of the Dialysis Department.

Summing up the analysis of surgical interventions for PD complications, it was noted that 13 (22.8%) patients underwent a surgical treatment without taking into account the catheter placement procedure itself. As for the removal of the catheter, in some cases it was planned, in some it was planned urgently, while in some cases it was unplanned when the decision to remove the catheter was made during the surgery. Two of the thirteen patients operated for PD complications were reimplanted with a catheter. One patient is discussed earlier in the paper (two catheter reimplantations with a difference of a day, then catheter removal), the second reimplantation was performed due to pericatheter leaks of dialysate.

Revision of the catheter required 19.3% of all receiving PD. The average surgical time was 50 min (±6 min), from 35–80 min maximum. Indications for the surgery were as follows: catheter dysfunction in eight (73%) patients, pericatheter leaking of dialysate in two (18%), and acute intestinal obstruction in one (9%). Discussing the last indication, it is worth noting that this is the second case of thin-intestinal intussusception for a child with ARF on PD. A three-year-old girl was admitted to the hospital with an HUS clinic, and indications were given for initiating PD, which was started on the same day. On the fifth day of dialysis, in connection with the acute intestinal obstruction clinic, the child was taken to the surgical room. During the operation, a thin-bowel invagination was diagnosed, the latter was straightened out, but the catheter was removed from the abdominal cavity, dialysis was stopped, and renal replacement therapy was discontinued.

The tactic of stopping dialysis for patients with PD complications was applied in eight cases (pericatheter leaks, after catheter revision, after peritoneal suturing, and after disinvagination), and in all cases was justified as it allowed first of all, to avoid additional intervention and, secondly, to achieve the normalization of biochemical parameters by conservative means.

CONCLUSION

PD has established itself as the first-line treatment for a renal replacement therapy for children with an ARF. Along with the growing popularity of the technique, the number of surgical complications has also increased [1, 11].

In the structure of surgical complications of acute PD are the leaders of noninfectious complications, including the dysfunction of the catheter and periuterine leaks. Dysfunction of the peritoneal catheter is in the first place among all surgical complications of PD for children with ARF.

Despite the performed resection of the omentum during implantation of the catheter in order to prevent its entanglement, a significant part of the dysfunction occurs due to obturation of its fat suspensions (up to 50%). Unfortunately, nowadays, there are no methods for preventing these complications. Given the high risk of reimplantation of the catheter during its revision through the main wound, we recommend performing the intervention with a separate access over the womb. Although the use of laparoscopy in the diagnosis and treatment of catheter dysfunction is extremely promising due to its cosmetic nature, it is limited due to the inability to create adequate tightness in the first hours after a surgery.

Infectious complications of PD (in particular peritonitis) in children with a short period of receiving this type of dialysis therapy are quite rare and are usually explained by the presence of concomitant complications.

Despite the high percentage of surgical interventions, conservative tactics are justified in many cases. In all cases of premature termination of PD, the transition to hemodialysis was not required.

A more detailed study of the problem of noninfectious surgical complications of PD will create a recommendation base for implantation and care of catheters for the prevention and treatment of these complications that will ultimately improve the results of a treatment of children with an acute kidney failure.

About the authors

Dmitrii A. Dobroserdov

North-Western State Medical University named after I.I. Mechnikov; Saint Petersburg Children’s Hospital No.

Author for correspondence.
Email: dimit@bk.ru

Assistant Professor Department of Pediatric Surgery

Russian Federation, Saint Petersburg

Mikhail V. Shchebenkov

North-Western State Medical University named after I.I. Mechnikov; Saint Petersburg Children’s Hospital No.

Email: shebenkovmihail@gmail.com

MD, PhD, Dr. Med. Sci. Professor Department of Pediatric Surgery

Russian Federation, Saint Petersburg

Alexey L. Shavkin

Saint Petersburg Children’s Hospital No. 1

Email: dialys.dgb@gmail.com

Associate Professor, Head Dialysis Department

Russian Federation, Saint Petersburg

References

Supplementary files