Long-term outcomes of the repeated lumbar microdiscectomies using polyacrylamide hydrogel: clinico-radiological relevance

Педаченко Є.Г.1, ORCID:0000-0003-4759-6019, e-mail: pedach@i.ua Хижняк М.В.1, ORCID: 0000-0002-6632-4206, e-mail: khyzhnyak63@gmail.com Красиленко О.П.1, ORCID:0000-0002-6936-6647, e-mail: elena.krasylenko@gmail.com Педаченко Ю.Є.1, 2, ORCID: 0000-0003-0609-424X, e-mail: yupedachenko@gmail.com Танасійчук О.Ф.1, ORCID: 0000-0003-3842-1786, e-mail: shepilo@3g.ua Крамаренко В.А.1, ORCID: 0000-0002-9662-0874, e-mail:vladimirkramarenko@ukr.net Фурман А.М.1, ORCID:0000-0002-1229-0821, e-mail: andrey-furman@ukr.net Макеєва Т.І.1, ORCID:0000-0002-0210-9320, e-mail:tatyana.makieieva@gmai.com Стулей В.А.3, ORCID: 0000-0001-7552-7024, e-mail: stuley.volodymyr@lll.kpi.ua Земскова О.В.1, ORCID: 0000-0001-9462-8330, e-mail: oxzemskova@gmail.com


Relationship with academic programs, plans and themes
The study was performed within the framework of the research project «To develop the methods for preventing adhesive epidural process in patients following lumbar microdiscectomies and to assess their efficacy ( affecting the residual radicular pain as EF, the displacement of spinal root at the postoperative site, the repeated recurrent IVD herniation the repeated recurrent IVD herniation and de novo IVD herniation. Results. In 12 months following the surgical treatment, EF frequency was significantly lower in MG as compared with CG where PAH was not used (p = 0.02936). Nevertheless, the difference between groups was not significant when MRI findings of the displacement of spinal root at the postoperative site (р = 0.46759), the frequency of the repeated recurrent IVD herniation (p = 0.90904) and de novo IVD herniation (p = 0.60385) were compared. In study group, there were no association between MRI signs of postoperative EF, the displacement of spinal root or new recurrent IVD herniation on the one hand and the frequency of the radicular pain localized at the area of the surgical intervention on the other hand (p = 0.66482; p = 0.09515; p = 0.22857, respectively). In CG, the significant association between the displacement of spinal root at the postoperative site and the clinical symptoms above was revealed (p = 0.00222) with association coefficient φ = 0.41 indicating the moderate strength of this association. Conclusions. The data obtained generally suggest the reasonability of applying PAH manufactured in Ukraine as a barrier material at least for the treatment of the patients with recurrent IVD herniation representing the challenging category of spinal patients.
Наші уявлення про механізм формування епідурального фіброзу (ЕФ) постійно поглиблюються, і натепер вивчено чимало складових багаторівневого каскаду патофізіологічних реакцій -від його пускових чинників до кінцевого результату [3]. Вважають, що поступове заміщення в епідуральному просторі жирової тканини фіброзною, запускається асептичним Herniation of intervertebral discs (IVD) with neurocompression syndrome represents the most frequent complication of the degenerative diseases of the backbone and requires the surgical treatment. Nevertheless, the surgical removal of IVD hernia does not always lead to the desired recovery. 10-40 % of patients are diagnosed with the failed back surgery syndrome manifesting in back pain and/or the signs of the compression or irritation of the nerve structures due to postoperative structural and biomechanical changes of the backbone [1]. Among the major factors leading to backbone surgery failure syndrome even after correct surgery according to the indications are the repeated sequestration of the element of nucleus pulposus, the instability of the segment having been operated, the progression of the degeneration of IVD and zygapophysial joints, and the coarse adhesive epidural process [2].
There is growing understanding of the mechanisms implicated in the formation of epidural fibrosis (EF). Nowadays, several components of the multilevel cascade of the pathophysiological reactions ranging from the triggers of the process to the final result have been clarified [3]. The progressive substitution of adipose tissue in the epidural space by the fibrous one is thought Ukrainian journal of radiology and oncology. 2021;29(4):9-25 ISSN 2708-7166 (Print) ISSN 2708-7174 (Online)
Мета роботи -оцінити радіологічні показники стану хребта (за даними МРТ) та клінічну симптоматику оперованих хворих, а також визначити силу їх взаємозв'язку у віддаленому періоді після повторних поперекових мікродискектомій з приводу рецидиву гриж МХД, виконаних із інтраопераційним епідуральним введенням ПГГ. to be triggered by the aseptic inflammation with underlying dynamic imbalance in the system of regulating molecules and corresponding changes of cells and extracellular matrix. The shift in the ratio between proinflammatory and anti-inflammatory cytokines is accompanied with the sequential infiltration of the wound with the macrophages of the functionally different phenotypes [4]. The increase in proinflammatory cytokines (TNF-α, IL-6) is correlated positively with the initial accumulation of M1 macrophages [5]. The following increase in the level of proinflammatory (TGF-β, IGF-1) and profibrotic (in particular, ARG-1) factors is associated with the activation of M2 macrophages [6]. Macrophage transformation is important for the broad spectrum of the sequential processes of wound healing -in clearing, revascularization and re-epithelization as well as scarring via stimulation of fibroblast proliferation, differentiation of myofibroblasts and collagen deposition. There is considerable evidence that the changes in macrophage numbers, their density and metabolism contribute to the impairment of the above processes facilitating excess scarring [4,7]. Fibroblasts are critical players in fibrotization processes due to production of large amounts of collagen and fibronectin at the site of injury [8][9][10][11][12][13]. The experimental studies clarify several specific mechanisms of EF formation realized via neutrophil extracellular traps (NET), signal transduction pathways HMGB1/TLR4 and TGF-β1/Smad3, etc. [14,15].
Several factors affecting the intensity and prolongation of the aseptic inflammation are known. The severity of the aseptic inflammation is correlated with the volume of the intraoperative hematoma while its long-term duration is maintained by the autoimmune response to IVD elements appearing in the spinal canal that possess antigenic properties as an extra barrier tissue [16][17][18]. Some technical features of the operationcontributing to its overall traumatic effects are alsoimportant [19].
The proved pathogenetic elements and factors are now considered as the promising candidates for correction aimed at preventing EF [20]. Currently, various conservative (invasive and non-invasive) and surgical approaches are used for EF prevention [21][22][23][24][25][26]. Many years of experience in the usage of different barrier materials in microdiscectomy such as polyacrylamide hydrogel (PAH) have been accumulated. Nevertheless, the results of such treatment are rather controversial [27][28][29][30][31].
Polyacrylamide hydrogel (PAH) manufactured in Ukraine is an example of barrier material possessing the acceptable safety profile and used in clinical practice [32].
Nevertheless, it has not yet been clarified to which extent the barrier materials used after microdiscectomy could modify the development of EF. The EF effect on the clinical outcome in spinal surgery is also worth studying. Our study focused on these issues.
Purpose -of the study was to assess the radiological data on the state of the backbone according to MRI findings and the clinical symptoms of the patients following the surgery as well as the determine the long-term strength of the interaction between the factors under study after the repeated lumbar microdiscectomies performed with intraoperative epidural PAH administration. Ukrainian journal of radiology and oncology. 2021;29(4):9-25 ISSN 2708-7166 (Print) ISSN 2708-7174 (Online)
The results of the clinical and radiological assessment of 96 patients enrolled into the single-center prospective cohort study performed at the State Institution «Romodanov Neurosurgery Institute, National Academy of Medical Sciences of Ukraine» have been presented. All patients were operated in 2017-2020 in the clinic of minimally invasive and laser spinalneurosurgery with X-ray operating room.
The main group (MG) included 35 (36.5 %) patients with recurrent IVD herniation to whom PAH was administered epidurally at the final step of the surgery comprising disc removal and decompression of the nervous structures with the aim of preventing cicatrical adhesive EF. In control group (61 patients, 63.5 %), the removal of the recurrent IVD hernia and decompression of the spinal root was not followed by the administration of the barrier materials were not used.
Inclusion criteria: -age of patients from 18 to 76 years; -late postoperative period (12 months) after repeated lumbar microdiscectomy due to the recurrent IVD herniation; -voluntary informed written consent to participate in the study approved by the Committee on Ethicsand Bioethics of the State Institution «Romodanov Neurosurgery Institute, National Academy of Medical Sciences of Ukraine» (Meeting Minutes No.3 of June 6, 2016).
Exclusion criteria: -age of patients below 18 and above 76 years; -postoperative period less than 12 months after repeated lumbar microdiscectomy due to the recurrent IVD herniation; -functional status of the patient according to Karnofsky scale is less than 70 points; -comorbid diseases at the stage of decompensation; case history of cancer; -pregnancy; -patient unwillingness to participate in the study, unsatisfactory compliance.
The clinical characteristics being studied are given in Table 2.
Частотні характеристики вибірки щодо зареєстрованих подій для досліджуваних факторів, розраховані по відношенню до чисельності хворих у відповідних групах, наведено на рис. 1. Table 2, all patients had the radicular pain prior to the surgery. This radicular pain was the major manifestation of the neurocompression syndrome being the indication for preoperative selection. In both groups of the study, Oswestry disability index (ODI) [33] reflecting the overall self-assessment of the quality of life taking into account the restrictions due to the pain corresponded to the severe disability.

As shown in
The mean score of IVD degeneration according to five-point Pfirrmann grading system [34] in MG was 3. The frequency distribution of the registered events for the factors being studied calculated relative to the number of patients in corresponding groups is given in Fig. 1.
The data presented in Fig. 1 demonstrate that in 12 months following the surgery, the signs of EF and the spinal root displacement in MG were less frequent than in CG. Nevertheless, the difference in the rate of the residual radicular pain with localization matching the site of the surgical intervention in MG and CG was less pronounced. This empirical analysis allowed us to formulate the major hypothesis of the study that were verified with the aid of the corresponding statistical tests stated below: The primary end-points of the study were the long-term outcomes of the repeated lumbar microdiscectomies: EF rate at the postoperative site; the rate of the displacement of spinal root at the postoperative site; the rate of the repeated recurrent IVD herniation and de novo IVD herniation; the rate of the cases with the residual radicular pain localized at the dermatome matching the surgical intervention site and the extent of the worsening of life quality in the treated patients.
The secondary end-points of the study were the following: the analysis of the associations with delineation of the strength of such factors affecting the residual radicular pain as EF, the displacement of spinal root at the postoperative site, the repeated recurrent IVD herniation the repeated recurrent IVD herniation and de novo IVD herniation.
EF and the displacement of spinal root at the postoperative site, presence/absence of the new recurrence or de novo IVD herniation were assessed by MRI of the lumbar spine in 12 months following the surgery on «Intera 1.5 Tl» (Philips, Netherlands) with a magnetic field induction of 1.5 T. MRI examination protocol comprised the following sequences: sagittal and axial T1w_TSE and T2w_TSE standard turbo spin echo, sagittal with suppression of MR signal from adipose tissue (STIR_TSE), coronal T2w_TSE, sagittal and axial T1w_TSE with intravenous administration of gadolinium compounds as paramagnetic.
In our study, the critical significance level I was taken as α = 5 % (the probability of rejecting the null hypothesis (Н 0 ) when it is true). The calculated p-values were compared with the standardized 5 % level. The null hypothesis was rejected when p-value ≤ 0.05.
For analyzing quantitative variables and verifying group homogeneity, two independent groups were compared by Mann-Whitney non-parametric U test.
The categorical variables were compared in c 2 -test and two-tailed Fisher exact test as to the expected frequencies in the contingency tables.

Analysis of group homogeneity
The statistical homogeneity of the patients' groups for the quantitative variables was checked by Mann -Whitney non-parametric U test. According to the results of the calculation, the statistic hypothesis H0 was accepted as to the absence of significant preoperative differences between the groups in age, BMI, the grade of IVD degeneration (by Pfirrmann system) and the clinical status based on the self-assessment of the quality of life (ODI) and the presence of the residual radicular pain (Table 3).
The gender distribution of the patients in MG and CG was compared using c 2 -test (PearsonChi-square).
The statistically significant difference was not demonstrated (p >0.05): (df = 1; p = 0.82401). Therefore, H 0 hypothesis was accepted as to the absence of non-homogeneity in gender distribution in the compared groups.
Nevertheless, the calculated number of the expected cases of postoperative EF in MG (8.4) was more than twice higher than actual number. In CG, the calculated number of the expected cases of postoperative EF was 26.6 as compared to 31 actual EF cases (Table 5).
Consequently, in the cohort under study, interaoperative epidural administration of PAH is associated with the significant decrease in the frequencies of the radiological signs of EF detected in 12 months following the surgical treatment of the recurrence of IVD herniation.
The MRI signs of the displacement of spinal root at the postoperative site were observed in 9 (25.7 %) patients of MG and 20 (32.8 %) patients of CG (to whom barrier materials were not administered) (see Table 4). The ratio of calculated expected to actual cases of the displacement of spinal root was 10.6:9 in MG and 18.4:20 in CG, i.e. this ratio in the compared groups was opposite.
At the same time, the difference between groups as to the frequency of MRI signs of the displacement of spinal root was not significant (df = 1; p = 0.46759), contrary to that in EF frequencies. Therefore, referring to this sign, H 0 has been accepted.
The MRI signs of the new recurrence of IVD herniation in 12 months following the repeated microdiscectomy were detected in 1 (2.9 %) case in MG and 2 (3.7 %) cases in CG (Table 4). Besides, in 2 (5.7 %) cases in MG and 3 (4.9 %) cases in CG, de novo IVD hernia was revealed ( Table 4). The difference between MG and CG concerning these two signs also did not reach the level of the statistical significance (df = 1; p = 0.90904 and df = 1; p = 0.60385, respectively).
In 12 months following the repeated surgery, the residual radicular pain with localization matching the area where IVD was removed was registered in 8 (22.9 %) patients of MG and 15 (24.6 %) patients of CG (Tables 2 and 4). Considering this symptom, the difference between the groups was not significant (df = 1; p = 0.84815). The residual radicular pain was rather mild (on average 2.00 ± 0.33 points in MG and 2.20 ± 0.20 points in CG according to 10-points visual analogous scale [36] (p = 0.477676)). In addition, the pain was significantly less pronounced (p < 0.005) as compared to the most intensive radicular pain prior to the repeated surgery (on average 9.13 ± 0.35 points in MG and 9.31 ± 0.22 points (p = 0.690599).
Meanwhile, it should be noted that de novo hernias visualized in 12 months following the surgery in the intact spinal motion segments or contralateral segments at the level of surgical intervention have been manifested with the radicular pain in neither group under study.
(7.37 ± 0.83 % in MG and 8.20 ± 0.71 % in CG) ( Table 2). The difference of these values between the groups was not statistically significant. Meanwhile, the empirical analysis of the data presented in Fig. 3 allows us to suggest that less impairment of the life quality seems to be registered when PAH is administered.
However, this suggestion should be verified in more numerous samples since within the framework of this study H 0 have not been rejected.
За нашими даними, у хворих ОГ не зареєстровано статистично значущої асоціації між післяопераційним ЕФ та залишковим корінцевим болем, що відповідав зоні видалення рецидиву грижі МХД Assessment of association between the studied radiological and clinical features According to the objectives of this study, we analyzed the associations between the frequency of the residual radicular pain in 12 months after the surgery localized in the area matching the site of the surgical intervention and such factors as EF, the displacement of spinal root at the postoperative site, the repeated recurrent IVD herniation the repeated recurrent IVD herniation and de novo IVD herniation.
We were interested in the association between radiological findings and clinical symptoms. The data on this subject are still controversial. Several authors demonstrated that the residual radicular pain in cases of the extensive EF is thrice more frequent than in cases with less pronounced cicatrical adhesions [37]. However, there are data about asymptomatic EF [38,39].
According to our data, neither patients in MG (p = 0.66482), nor in CG (p = 0.22940) demonstrated significant association between postoperative EF and the residual radicular pain localized in the area matching the site of the surgical intervention. Ukrainian journal of radiology and oncology. 2021;29(4):9-25 ISSN 2708-7166 (Print) ISSN 2708-7174 (Online)
Nevertheless, some differences between MG and CG have been proved when association between other morphological signs and clinical symptoms was analyzed. In particular, in MG no association was demonstrated between the displacement of the spinal root at the postoperative site and the residual radicular pain of the corresponding localization (p = 0.09515). As anticipated, the same holds true for frequencies of recurrent IVD or IVD herniation de novo that were not associated with the residual radicular pain of the corresponding localization (p = 0.22857 and p = 0.41008, respectively).
In CG group, the situation was reversed. Namely, there was a statistically significant association between the frequency of MRI-visualized displacement of the spinal root in postoperative area and the radicular pain in corresponding dermatome (p = 0.00222). The association coefficient (φ = 0.41) indicates the moderate strength of such association. Also, in CG, the trend towards the impact of de novo recurrence of IVD herniation on the frequency of the clinical manifestations stated above was observed (p = 0.05738). However, this statistical hypothesis requires to be verified in most powerful samples.
To generalize this part of the study, we should note that in 12 months following the repeated surgery the presence of MRI-visualized EF features at the level of surgical intervention did not affect significantly the frequency of the residual radicular pain of the corresponding localization irrespective of intraoperative administration of PAH. It should be stressed that such pattern was evident in the setting of the statistically significant decrease of the frequency of MRI features of EF in MG.
In MG group, the displacement of the spinal root at the postoperative site or de novo IVD herniation did not affect significantly the frequency of the residual radicular pain of the corresponding localization. On the contrary, in CG where PAH was not administered, the moderate strength association between this clinical symptom and MRI signs of the displacement of the spinal root at the postoperative site was evident. In addition, the trend towards the impact of de novo recurrence of IVD herniation on the frequency of these clinical manifestations was observed. At the same time, in spite of the differences in these associations, MG and CG did not differ significantly by the frequency of MRI signs of the displacement of the spinal root and de novo recurrences of IVD herniation.
2. Epidural administration of PAH did not affect significantly the frequency of the detection of the radiological signs of the displacement of the spinal root at the postoperative site, recurrence of IVD herniation or de novo IVD hernias in 12 months following the surgery.
5. In patients operated without PAH administration, the moderate strength association between the residual radicular pain and MRI signs of the displacement of the spinal root at the postoperative site was evident. In addition, the trend towards the impact of de novo recurrence of IVD herniation on the frequency of this clinical symptom was observed. At the same time, in this group, no significant association between MRI signs of EF and the frequency of the residual radicular pain was observed.
The data obtained testify to the expediency of using PAH produced in Ukraine as the barrier material, at least, in the difficult category of spinal patients with recurrent IVD herniation. Nevertheless, it should be stated that our analysis concerned each factor taken separately. Therefore, more generalized conclusions as to the overall effect of the epidural PAH administration on the outcomes of the repeated microdiscectomies should be drawn with care.