*2.2. Cytotoxicity Testing*

Cytotoxicity testing was performed using the method described in the ISO-10993-5 guideline. Accordingly, the specimens were divided into the following four groups: test (ES PLA95/β-TCP), reagent blank control (medium), negative control (HDPE material), and positive control (zinc sulfate). The samples were extracted with Eagle's minimal essential medium (α-MEM; GIBCO BRL, OK, USA) containing 10% fetal bovine serum (GIBCO BRL, USA) at 37 ± 1 ◦C for 24 ± 2 h. Cell line (NCTC clone 929; ATCC) was cultured in each of the extraction medium, with 5% CO2 at 37 ◦C for 48 h (*N* = 3). A light microscope was used for qualitative morphological grading of the cytotoxicity test findings.

#### *2.3. In Vivo Test (Animal Model)*

Four healthy LanYu pigs (weight: 20–25 kg) were used for animal studies. The protocol was approved by Taipei Medical University (No. LAC-99-0087). Buccal mucoperiosteal flaps were reflected in the bilateral mandibular premolar and molar areas. Buccal alveolar bone was reduced to a level 5-mm apical to the cement–enamel junction (CEJ). The root surface was denuded of the periodontal ligament (PDL) and cementum (CE), and notches were placed at the bone level of each root as in Figure 2. The ES PLA95/β-TCP and control membranes were placed on individual bone defect areas without bone grafting. Flaps were positioned and sutured. All LanYu pigs were sacrificed at the designated times after surgery. Histological and histometric evaluation at 8 and 16 weeks were performed after surgery respectively, to determine the healing response of each treatment modality. Hematoxylin and eosin stain (H&E) staining of the demineralized animal sections were evaluated under a light microscope (40×), and the CE and bone height were measured using the Image J software (NIH, MD, USA).

**Figure 2.** From left to right: Animal model for guided tissue regeneration (GTR) membrane; representative surgery photo and overview of the animal study: the GTR membranes of ES PLA95 and ES PLA95/β-TCP were experimental groups, while the Epi-Guide and empty defect site were used as control groups. Histological and histometric evaluation at week 8 and 16 were performed to determine the healing response of each modality.
